Late Quaternary Volcanic Stratigraphy Within a Portion of the Northeastern Tongariro Volcanic Centre

Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author.

LATE QUATERNARY VOLCANIC STRATIGRAPHY WITHIN A PORTION OF THE NORTHEASTERN TONGARIRO VOLCANIC CENTRE

A thesis presented as partial fulfilment of the requirements for the degree of

Doctor of Philosophy in Soil Science

Shane Jason Cronin

Massey University Palmerston North New Zealand

1996 Mount Ruapehu viewed from the east, July 1996. Fresh tephra covers snow on the northern sector of the volcano.

ii ACKNOWLEDEGMENTS

I thank Vince Neall, Bob Stewart and Alan Palmer for all their help during the course of this study. The enthusiasm and passion they have for their science has taught me that geology is not just a job, nor a discipline, but a lifelong process of discovery.

In particular I thank Vince for "letting me loose" on this project, even if it seemed that I was heading in the wrong direction at times.

Cleland Wallace, John Kirkman, Brent Alloway and David Lowe have all provided valuable and highly appreciated input into various parts of this work, as have journal reviewers listed within the thesis.

I was the grateful benefactor of funding from the New Zealand Vice Chancellors Committee, the Helen E. Akers Scholarship fund (twice!), the Massey University Graduate Research Fund and the Tongariro Natural History Society.

Travel funding from the Royal Society of New Zealand (Young Scientists' Fund), Massey University Research Fund, and the New Zealand Vice Chancellors Committee (Ciaude McCarthy Fellowship) enabled me to attend two international conferences, which has greatly enhanced my Ph.D. experience.

I am grateful to Justlands, the Department of Conservation and the Rangipo Forest Trust for access into parts of the study area.

I thank all my contemporaries in the Department of Soil Science for making it such a relaxed and "down to earth" place to work, particularly John (pH) Morrell and Shivaraj (Bucket) Gurung.

Last but not least, I thank my fa mily and Iris for their patient support during this work.

iii ABSTRACT

Investigation of the Late Quaternary volcanic stratigraphy within the andesitic Tongariro Volcanic Centre has elucidated the history of construction of the northeastern Ruapehu and eastern Tongariro ring plains and provided a lahar record for the Tongariro catchment. Volcaniclastic ring-plain sequences were correlated and dated using rhyolitic and andesitic marker tephras. The identification of distal rhyolitic tephras in the area was improved by the application of discriminant function analysis (DFA) to their electron microprobe determined glass chemistry. The Okaia, Omataroa and Hauparu Tephras and the Rotoehu Ash were identified for the first time in this area, providing a chronology for pre- 22.6 ka ring-plain sequences not previously investigated. DFA of ferromagnesian mineral chemistry proved useful for discrimination of andesitic tephras, with titanomagnetite being the most useful phase. Development of an andesitic tephrostratigraphy in pre-22.6 ka sequences was aided by clustering analysis and DFA. Seven andesitic marker tephras were identified using a range of parameters to supplement the rhyolitic tephrostratigraphy. Using the tephrochronologic framework, 15 packages of lahar deposits were identified on the northeastern Ruapehu ring plain (from >64 to c. 5.2 ka) and six on the eastern Tongariro ring plain (from >22.6 to 11.9 ka). Lahar deposition on both ring plains was most voluminous and widespread during the last (Ohakean) and antepenultimate (Porewan) stadials of the last glacial (Otiran). Holocene lahars were restricted to a narrow sector of the northeastern Ruapehu ring plain. They appear to have been triggered mostly in response to large-scale tephra eruptions of Ruapehu and Tongariro, and mostly occurred along the path of the Mangatoetoenui Stream. Lahar deposits and surfaces beside the Tongariro River were mapped in eight lahar hazard zones, with lahar recurrence intervals ranging from 1 in >15 000 years to 1 in 35 years. The largest number and volume of lahars in this catchment occurred in the period from 14.7 to 10 ka. The greatest population risk identified in the Tongariro catchment is part of Turangi, built within a 1 in 1000 year lahar-hazard zone. Other property and infrastructure at greater risk include the State Highway 1 bridge across the Mangatoetoenui Stream and the Rangipo Dam and Power Station, within a 1 in 35 year hazard zone. The landscape of the northeastern Ruapehu and eastern Tongariro ring plains has developed in relation to late Quaternary climate changes in addition to volcanic activity. During the last and antepenultimate stadials of the last glacial, major ring-plain aggradation by lahars and streams occurred. This was probably in response to greater physical weathering and glacier action on the volcanic cones providing abundant sediment for lahars. During the warmer interstadials of the last glacial, soil development within andesitic ring plain material was greatest, particularly when the rate of soil accretion was low.

iv TABLE OF CONTENTS

PREFACE

Frontispiece ...... ii Acknowledgements iii Abstract iv Table of contents r V

s l CHtl�HAPH��JTER;: ONE:. . . . .IN . TRODUC...... TION, . . . . . OBJECTIVES,...... OUTLI . . . . . NE. . . OF. . . STUDY...... AREA. . . . . AND� PREVIOUS WORK

- 1.1 Introduction ...... 1 1.2 Objectives of study 1 1.3 Methods . 1 - 1.4 Geography of the study area 2 -1.5 Climate, soils and land use 4 -1.6 Regional geologic setting 6 -1.7 Tongariro Volcanic Centre 10 1.8 Previous work in and around the study area 10 -1.9 Summary and conclusions from previous work 15 1.10 References ...... 18

CHAPTER 2: RHYOLITIC TEPHROCHRONOLOGY

2.1 Introduction ...... 21 2.2 Additional methodology and notes 22 .... 2.3 Methods of identifying late Quaternary rhyolitic tephras on the ring 22 plains of Ruapehu and Tongariro volcanoes, New Zealand 2.4 Keywords 2.5 Introduction 2.6 Setting 2.7 Methods 2.7.1 Sample preparation and analysis 2.7.2 Statistical methods - 2.7.3 Tephra classification methodology 2.8 Results 2.8.1 Ferromagnesian assemblages 2.8.2 Glass chemistry and similarity coefficients 10-22 ka tephras 22-65 ka tephras 2.8.3 Canonical discriminant function analyses of glass analyses 10-22 ka tephras discriminant model Classification of 10-22 ka unknown tephras 22-65 ka tephras discriminant model Classification of 22-65 ka unknown tephras

V CHAPTER 3: ANDESITIC TEPHROCHRONOLOGY

3.1 Introduction ...... 41 3.1.1 Photographs of the Upper Waikato Stream sequence 41 3.2 Contributions of co-authors 44 3.3 Sourcing and identifying andesitic tephras using major oxide 45 titanomagnetite and hornblende chemistry, Egmont volcano and Tongariro Volcanic Centre, New Zealand 3.3.1 Introduction 45 3.3.2 Methods 48 3.3.3 Statistical methods 48 3.3.4 Discrimination between sources 49 Titanomagnetite 49 Hornblende 50 3.3.5 Discrimination of individual Egmont volcano-sourced tephras 52 Ti tanomagnetite 52 Hornblende 53 3.3.6 Summary and conclusions 55 3.3.7 Acknowledgements 56 3.4 A multiple-parameter approach to andesitic tephra correlation, 56 Ruapehu volcano, New Zealand - 3.4.1 Introduction 57 3.4.2 Setting 57 3.4.3 Methods 59 3.4.4 Statistical methods 59 3.4.5 Stratigraphy and distribution of 23-75 ka andesitic tephras 61 3.4.6 Physical properties and mineralogy of 23-75 ka andesitic tephras 63 3.4.7 Mineralogy 64 3.4.8 Mineral chemistry of 23-75 ka andesitic tephras 66 3.4.9 Titanomagnetite (discrimination of Marker Units 1, 2 and 6) 67 3.4.10 Hornblende (discrimination of Marker Unit 4) 71 3.4.11 Olivine (discrimination of Marker Unit 5) 73 - 3.4.12 Other phases 74 "' 3.4.13 Conclusions 75 3.4.14 Acknowledgements 76 3.5 Combined reference list ...... 77

CHAPTER 4: PALEOSOL DEVELOPMENT AND PALEOCLIMATIC INVESTIGATIONS OF THE RING PLAIN SEQUENCE

4.1 Introduction ...... 81 4.2 Investigation of an aggrading paleosol developed into andesitic ring 82 plain deposits, Ruapehu volcano, New Zealand. 4.3 Introduction 82 4.4 Setting 83 4.5 Stratigraphy 83 -4.6 Physical description 84 4.7 Mineralogy of ash grade material 86 4.8 M��s M 4.9 Results and discussion 88 4.9.1 Primary ash mineralogy 88 4.9.2 Siliceous phases 88 4.9.3 Secondary minerals 90 - 4.10 Relationship of Ruapehu ring-plain sequence to climate record for 94 southern North Island

vi 4.11 Conclusions 96 4.12 Acknowledgements 97 4.13 References ...... 98

CHAPTER 5: THE LAHAR RECORD AND CONSTRUCTION OF THE NORTHEASTERN RUAPEHU AND EASTERN TONGARIRO RING PLAINS

5.1 Introduction ...... 101 5.2 A late Quaternary stratigraphic framework for the northeastern 102 Ruapehu and eastern Tongariro ring plains, New Zealand 5.3 Keywords 102 5.4 Introduction 102 5.5 Setting 103 5.6 Time control 103 5.7 Terminology and identification of ring plain deposits 103 5.8 Present channel geography 106 .9 Group 1 - Northeastern Ruapehu streams 107

5.10 Group 2 - Streams draining both volcanoes 110 �5.11 Group 3-Eastern and northeastern Tongariro streams 111 .12 Lahar distribution on the eastern ring-plain sectors from 23 ka to the present 113 - 5.13 Source and generation of lahars 118 , 5.14 Summary and conclusions 122 5.15 Acknowledgements 122 5.16 References ...... 123

CHAPTER 6: LAHAR HISTORY AND LAHAR HAZARD OF THE TONGARIRO RIVER

6.1 Introduction ...... 125 6.2 Lahar history and lahar hazard of the Tongariro River, northeastern 126 Tongariro Volcanic Centre, New Zealand 6.3 Keywords 126 6.4 Introduction 126 6.5 Setting 127 6.6 Terminology 127 > 6.7 Hazards of lahars 129 6.8 Methods 130 6.9 Lahar history of the Tongariro River 130 6.9.1 Stratigraphic record within the tributary streams 130 6.9.2 Stratigraphic record within the Tongariro River valley 131 6.9.2A Highest surface 131 6.9.28 Lower surfaces 134 6.10 Lahar hazards in the Tongariro River 136 6.1 0.1 1 in >15 000 year zone 137 6.1 0.2 1 in 12 000 year zone 137 6.10.3 1 in 10 000 year zone 137 6.1 0.4 1 in 5 000 year zone 139 6.1 0.5 1 in 2 000 year zone 139 6.1 0.6 1 in 1 000 year zone 139 6.10.7 1 in 100 year zone 139 6.1 0.8 1 in 35 year zone 141 6.11 Discussion 141 6.11.1 Mitigation of hazards 142 6.12 Conclusions 143 6.13 Acknowledgements 144 6.14 References ...... 144

vii CHAPTER 7: GEOLOGY AND LANDSCAPE DEVELOPMENT OF THE NORTHEASTERN TONGARIRO VOLCANIC CENTRE

- 7.1 Introduction ...... 147 - 7.2 Surficial geologic map of the northeastern Tongariro Volcanic Centre 147 7.2.1 Methods 147 7 .2.2 Laharic surfaces 15 0 7.2.2A Ruapehu-derived lahar surfaces 15 0 7.2.28 Tongariro-derived lahar surfaces 15 1 7.2.3 Lavas 15 1 7.2.3A Ruapehu-derived lavas 15 1 7.2.38 Tongariro-derivedla vas 15 3 7 .2.4 Moraines 15 4 7.3 Geological history of the northeastern ring plain of Ruapehu volcano, 15 4 New Zealand - 7.3.1 Introduction 155 - 7.3.2 Geologic setting 155 7.3.3 Site of study 15 6 I' 7.3.4 Stratigraphy 15 6 7.3.4A Age 15 6 7.3.48 Sequence 15 7 7 .3.5 Lithology 160 7.3.5A Diamictons 160 - 7.3.58Ande sitic tephra 161 7.3.6 Interpretation of geological history 162 7.3.7 Discussion and conclusions 166 7.3.8 Acknowledgements 166 7.3.9 Subsequent comments on the paper 167 7.4 Summary geological synthesis of the northeastern Tongariro 167 Volcanic Centre 7.4.1 c. 75 000 - 64 000 years ago 167 7.4.2 64 000 - 36 000 years ago 168 7.4.3 36 000 - 23 000 years ago 169 7.4.4 23 000 - 15 000 years ago 169 7.4.5 15 000 -9 700 years ago 170 7.4.6 9 700 -250 0 years ago 171 7.4.7 25 00 - present 171 7.5 Combined references ...... 171

CHAPTER 8: CONCLUSIONS

8.1 Fulfilment of study objectives - Tephrochronology groundwork ...... 175 8.2 Fulfilment of study objectives - Specific objectives 176 8.2.1 An im proved assessment of the lahar hazards from Ruapehu 176 volcano 8.2.2 An assessment of the lahar hazards on the eastern Tongariro 176 rin g plain and the Tongariro River - 8.2.3 A better understandin g of the eruptive history(pa rticularly 176 that of tephra eruptives) of the two volcanoes 8.2.4 An overallsyn thesis of the landscape development of the 177 northeasternRuapehu and eastern Tongariro ring plains 8.3 Additional contributions made by this study 177 8.3.1 Rhyo/itic tephra identification 177 8.3.2 Andesitic tephra discrimination 178

viii 8.3.3 Halloysite/allophane formation in andesitic te phra s 178 8.3.4 Relationship of rin g plain construction to late Quaternary 179 climate change 8.4 Potential future work 179 8.5 References ...... 180

APPENDICES

APPENDIX 1: SAS PROGRAMS USED IN THIS STUDY

1.1 Programs ...... A 1 1.2 Reference A3

APPENDIX 2: RHYOLITIC GLASS ANALYSES ...... A4

APPENDIX 3: ANDESITIC TEPHRAS

3.1 Ferromagnesian mineral assemblages ...... A 11 3.2 Hornblende analyses A 13 3.3 Olivine analyses A 15 3.4 Titanomagnetite analyses A 18 3.5 Analyses of other phases A24 3.5.1 Chromite A24 3.5.2 Clinopyroxene A24 3.5.3 Orthopyroxene A24 3.5.4 Andesitic glass A25 3.5.5 Plagioclase A26 3.5.6 llmenite A26 3.5.7 Spine! A26

APPENDIX 4: SELECTED FIELD SECTION DESCRIPTIONS ...... A27-A71

ix j

LIST OF TABLES

Table 1.1 Stratigraphy of Tongariro Subgroup tephras and interbedded tephras 13 described by Topping (1973, 1974) and Topping and Kohn (1973). 1.2 Stratigraphy of lahar formations on the SE Ruapehu ring plain, 15 from Oonoghue (1991) and Purves (1991 ). 1.3 Stratigraphy of andesitic and rhyolitic tephras on the SE Ruapehu 16 ring plain, from Oonoghue et al. ( 1995). 2.1 Ages and sources of rhyolitic tephras within the two discriminant 27 models used in this study. 2.2 Ferromagnesian mineral assemblages (modal %) and location 28 unknown tephra samples on the Ruapehu ring plain. 2.3 Correlation matrix of similarity coefficients, comparing glass major 29 oxide chemistry of rhyolitic tephras found in this study with published glass chemistry (Stokes et al., 1992) from potential correlative tephras aged 10-22 ka. 2.4 Correlation matrix of similarity coefficients, comparing glass major 30 oxide chemistry of rhyolitic tephras found in this study with published glass chemistry from potential correlative tephras aged 22-65 ka. 2.5 Mean electron microprobe glass compositions of tephras used 31 in the initial 10-22 ka discriminant model. 2.6 02 values and classification efficiencies of the initial discriminant 33 model for the 10-22 ka tephras. 2.7 Probabilities of classification of unknown tephras in the 10-22 ka 33 range with their potential correlatives, using the initial 10-22 ka tephra discriminant model. 2.8 02 values and classification efficiencies of the updated discriminant 34 model for the 10-22 ka tephras. 2.9 Probabilities of classification of unknown tephras (that were poorly 35 classified by the initial model) in the 10-22 ka range with their potential correlatives, using the updated 10-22 ka tephra discriminant model. 2.10 Mean electron microprobe compositions of tephras used in the 35 23-65 ka discriminant model. 2.11 02 values and classification efficiencies of the discriminant model 36 for the 22.5-65 ka tephras. 2.12 Probabilities of classification of unknown tephras in the 22.5-65 ka 36 range with their potential correlatives, using the 22.5-65 ka tephra discriminant model. 3.1 Average titanomagnetite and hornblende chemistry from the two 49 tephra sources. 3.2 Classification efficiency of the between-source discriminant functions. 52 3.3 Egmont-sourced tephras used in the discrimination study. 53 3.4 Classification efficiencies of the discriminant function analyses for 53 the individual Egmont-sourced tephras. 3.5 Rhyolitic tephras identified within the eastern ring plain sequence, 61 Ruapehu. 3.6 Mean electron microprobe analyses of final titanomagnetite, 66 hornblende and olivine groupings. 3.7 02 values between titanomagnetite groupings. 69 3.8 02 values between hornblende groupings. 73

X 3.9 02 values between olivine groupings. 74 3.10 Summary of the criteria for identification of andesite marker tephras 75 in this study. 4.1 Rhyolitic tephra identified within the north-eastern ring plain sequence. 84 4.2 Soil physical properties of selected ash samples. 86 5.1 Tephra coverbeds used in this study. 105 5.2 Sedimentary features of ring plain deposits and their inferred mode 106 of deposition. 5.3 Summary of the eruptive activity of Ruapehu and Tongariro 119 volcanoes for the last 75 ka based on the ring plain tephra record (Topping, 1973; Donoghue et al., 1995; Cronin et al., 1996(a)). 6.2 Number and timing of lahars in tributary catchments of the 132 Tongariro River, based on the exposed stratigraphic record in each catchment (Cronin and Neall, in press). 7.1 Mineralogy and locations of lava sampled from the northeastern 152 Tongariro Volcanic Centre. 7.2 Rhyolitic tephra identified within the northeastern ring-plain 156 sequence, Ruapehu. 7.3 Surface-flow sediment lithotypes within the northeastern Ruapehu 161 ring plain area, with their characteristic properties and inferred mode of deposition.

LIST OF PLATES

Plate 1.1 View of the eastern flanks of Ruapehu volcano across its ring plain. 11 1.2 View of Tongariro volcano from the southeast. 11 3.1 Part of the Upper Waikato Stream sequence located at T20/468102 42 and represented in Figs. 3.6, 4.1 and 5.3. 3.2 Marker Unit 1 tephra at T20/464098. 42 3.3 Marker Unit 2 tephra at T20/4691 00. 43 3.4 Marker Unit 3 tephra at T20/4681 02. 43

LIST OF FIGURES

Figure 1.1 Location of the study area (shaded) within the Tongariro Volcanic 3 Centre, central North Island, New Zealand. 1.2 Infrastructure and land ownership within the studied area. 5 1.3 Soils of the study area. 5 1.4 Vegetation/land use within study area. 7 1.5 Major elements of the Pacific-lndo-Australian Plate boundary through 7 New Zealand and the positions of the Taupe Volcanic Zone and the Tongariro Volcanic Centre. 1.6 Sketch map of the geology of the region surrounding the study area, 9 adapted from Grindley (1960) and Hay (1967). 1.7 Generalised geologic map of Ruapehu and related vents, adapted 14 from Hackett (1985).

xi 2.1 Location map of the eastern ring plains of Tongariro and Ruapehu 25 volcanoes. 2.2 (A) Plot of the first two canonical variates (Can 1 and Can 2) for 32 tephras comprising the initial 10-22 ka rhyolitic glass discrimination model. (B) Plot of FeO vs. CaO weight % within the rhyolitic glass of the 32 tephras comprising the initial 10-22 ka discriminant model. (C) Plot of the first two canonical variates (Can 1 and Can 2) for 32 tephras comprising the updated 10-22 ka rhyolitic glass discriminant model. 2.3 Plot of the first two canonical variates (Can1 and Can2) for tephras 32 comprising the 22-65 ka rhyolitic glass discriminant model. 3.1 North Island of New Zealand with locations of Egmont volcano and 46 Tongariro Volcanic Centre. 3.2 Plots of the first and second canonical variates (Can1 and Can 2) 51 for titanomagnetite data, together with 02 values between source groups. (A) Sample mean data with all oxide variables, (B) sample mean data excluding Cr203, (C) all individual analyses using all oxides. 3.3 Plots of the first and second canonical variates (Can1 and Can2) 51 for hornblende data together with 02 values between source groups. (A) Sample mean data, (B) all individual analyses. 3.4 Plots of the first and second canonical variates (Can1 and Can2) for 54 Egmont-sourced tephras together with 02 values between tephra groups. (A) Using titanomagnetite data, (B) using hornblende data. 3.5 Location map of the eastern ring plains of Tongariro and Ruapehu 58 volcanoes. 3.6 Composite stratigraphic sequence from principal reference sections 62 observed in the Upper Waikato Stream area, with positions of andesitic and rhyolitic marker tephras shown. 3.7 Partial andesitic tephra isopach maps, (A) Marker Unit 1, 65 (B) Marker Unit 3. 3.8 Plot of Ti02 wt% vs. FeO(recalculated) wt% of titanomagnetite data. 65 3.9 Plots of the first and second canonical variates (Can1 and Can2) 70 for titanomagnetite data. (A) Cluster-defined groupings, (B) mineralogical groupings, (C) mineralogical groupings with reduced variables (Cr203 omitted), (D) chemically defined groupings. 3.10 Plots of the first and second canonical variates (Can1 and Can2) 72 for hornblende data. (A) Cluster defined groupings, (B) refined cluster-defined groups. 3.11 Plots of the first and second canonical variates (Can1 and Can2) 72 for olivine data. (A) Cluster defined groupings, (B) mineralogical groupings, (C) mineralogical groupings with reduced variables. 4.1 Stratigraphic column of part of the northeastern Ruapehu ring-plain 85 sequence with paleosol development and its position within the overall ring plain sequence. 4.2 Grain-size distribution within the fine ash materials assessed using 89 the methods of Alloway et al. (1992). 4.3 Selected mineralogy of the fine ash sequence. 92

xii 4.4 The relationship of the northeastern Ruapehu ring-plain sequence 95 with deep sea 8018 isotope record, and southern North Island loess stratigraphy from the Rangitikei valley, Wanganui and Taranaki. 5.1 Location of Tongariro Volcanic Centre including Mts. Tongariro 104 Ngauruhoe and Ruapehu and the streams dissecting the eastern Ruapehu and Tongariro ring plains. 5.2 A sequence of selected measured sections in the direction of 1 08 stream flow in the Mangatoetoenui Stream. 5.3 Composite stratigraphic sections for Group 1 streams draining 109 the northeastern flanks of Ruapehu. 5.4 Stratigraphy of lahar deposition episodes on the northeastern 111 Ruapehu and eastern Tongariro ring plains. 5.5 Composite stratigraphic sections for Group 2 streams draining 112 the northeastern flanks of Ruapehu and southeastern flanks of Tongariro. 5.6 Composite stratigraphic sections for Group 3 streams draining 114 the eastern flanks of Tongariro. 5.7 Distribution of deposits from lahar episodes: (A) R1 1/T4, (B) R10, 115 (C) R09/T3 and the Hinuera Formation, (D) T2. 5.8 Distribution of deposits from lahar episodes: (A) R08/T1, (B) R07 116 and R06, (C) R05, (D) R04. 5.9 Distribution of deposits from lahar episodes: (A) R03 and R01 , 117 (B) R02. 6.1 Location of the Tongariro catchment, draining the northeastern 128 Tongariro Volcanic Centre in the central North Island of New Zealand. 6.2 Stratigraphy of sections preserved through the highest laharic 133 surface beside the Tongariro River. 6.3 Stratigraphy of sections through the lower laharic surfacesalongside 135 the Tongariro River. 6.4 Lahar hazard map of the Tongariro River catchment. 138 6.5 Enlargement of lahar hazard map for Turangi and its surrounds. 140 7.1 Map of the surficial geology of a portion of the northeastern 148 Tongariro Volcanic Centre. 7.2 Legend of geologic units mapped in Fig. 7.1 with their relationship 149 to previously mapped formations and tephra marker beds. 7.3 Location of the Tongariro Volcanic Centre including Mts. Tongariro, 158 Ngauruhoe and Ruapehu, and the streams dissecting the eastern Ruapehu and Tongariro ring plains. 7.4 Composite stratigraphic columns of the northeastern Ruapehu ring- 159 plain sequence below the regional marker horizon of the Kawakawa Tephra. 7.5 Comparison of the northeastern Ruapehu ring-plain landscape events 164 with deep sea 8018 isotope record, rhyolitic tephra marker beds, and southern North Island loess stratigraphy from the Rangitikei valley, Wanganui and Taranaki. 7.6 Summary diagram of the estimated rate (per ka) of lapilli-producing 168 sub-plinian eruptions of Ruapehu and Tongariro volcanoes for the last 75 ka.

xiii CHAPTER 1: INTRODUCTION, OBJECTIVES, OUTLINE OF STUDY AREA AND PREVIOUS WORK

1.1 Introduction

Ruapehu and Tongariro are the two largest and youngest Quaternary volcanoes of the Tongariro Volcanic Centre in the central North Island of New Zealand. Both volcanoes are surrounded by ring plains composed dominantly of volcaniclastic deposits and lava flows. The ring plains are volumetrically as large or larger than the cone they surround. The ring plains provide arguably the best geologic record of the activity of the two volcanoes, because the cone itself is generally exposed to more severe physical weathering leading to greater erosion and loss of the stratigraphic record, and also older deposits tend to be buried by later erupted lavas. Of human importance, the geologic record contained within the ring plains provides information on the rates and magnitudes of tephra eruptives and lahars from the two volcanoes which are of the greatest threat to life and property.

1.2 Objectives of study

The manifold objectives of this study all branched from the main aim of establishing a stratigraphy for the northeastern Ruapehu and eastern Tongariro ring plains. Once this stratigraphy was developed, investigations were made into the chronology and rates of tephra eruptions, lahars, lava flows and also soil development and landscape stability. Prior to this study the last detailed work in this area of the ring plains was that of Topping (1973, and 1974). Topping's study was limited to the younger part of the stratigraphic record (<22.5 ka B.P.) and mostly to the Tongariro ring plain. The outcomes of this study were to be; (1) an improved assessment of the lahar hazards from Ruapehu volcano in conjunction with the previous studies of Purves (1990), Donoghue (1991) and Hodgson (1993) on other sectors of the volcano, (2) an assessment of the lahar hazards on the eastern Tongariro ring plain and the Tongariro River, (3) a better understanding of the eruptive history (particularly that of pre-22.5 ka tephra eruptives) from the two volcanoes, and (4) an overall synthesis of the landscape development of the NE Ruapehu and E Tongariro ring plains.

1.3 Methods

The initial part of the study was carried out in the field. The ring-plain areas were traversed and detailed section descriptions taken of all important exposures found. Effort was concentrated on the channels of rivers and streams which incise into the ring-plain deposits, affording the best exposure of the ring plain stratigraphy. Aerial photography was used where possible in conjunction with the field information to map deposits and/or surfaces using their coverbed stratigraphy. Previously dated tephra layers, both rhyolitic

1 and andesitic, were quickly found to be the most useful means of dating surfaces and deposits. Laboratory studies concentrated on the identification of these tephras and also on finding new marker tephra layers, particularly andesitic tephras. Laboratory studies involved analysing the mineralogy of tephras as well as major element mineral and glass chemistry using the electron microprobe technique. Statistical methods were developed and used for both rhyolitic and andesitic tephra identification from the chemical analyses. Further laboratory studies were also conducted on buried soil materials to investigate former soil-forming, landscape and paleoclimatic conditions.

1.4 Geography of the study area

This study was confined to the NE Ruapehu and E Tongariro ring plains and is effectively the volcanic portion of the Tongariro River catchment (Fig. 1.1 ). The southern extent of the area lies along the watershed boundary between the Whangaehu River to the south and the Tongariro River to the north. Donoghue (1991 ) studied the Ruapehu ring plain in detail south of this boundary. The studied area encompasses the full eastern extent of the two ring plains which terminate against foothills of the Kaimanawa Mountains in the approximate position of the Tongariro River channel. The Round-the-Mountain Track on Ruapehu and a similar altitude (ea. 1300 m) on Tongariro marks the western boundary of study. North of State Highway 47A the study area narrowed to the general vicinity of the Tongariro River and its terraces. The town of Turangi is the largest population centre within the area with a population of 4239 in 1991 (Department of Statistics, 1992). In addition, two smaller settlements, Rangipo and Tokaanu, and the Rangipo Prison lie within the study area. These settlements and other scattered dwellings are confined to the northern part of the area, while in the southern portion there are no permanent dwellings (Fig. 1.2). The Turangi-Tokaanu area has been occupied by Maori of the Ngati Tuwharetoa tribe since the 16th century and European settlement began in the Tokaanu area as trout were released into the Tongariro River in the early 1900s (Ciarke and Smith, 1986). Prior to 1964 less than 500 people lived in Turangi. The population swelled up to 7000 during the period of construction of the Tongariro Power Development from 1964 to 1983 (Mercer, 1973). Following completion of the Tongariro Power Development the population has dwindled and the town has become a service centre for a developing tourist industry. In addition, the town services extensive exotic forestry planting and the surrounding farming industry. The southern part of the study area contains a large portion of the Tongariro National Park, while other parts of the area are Ngati Tuwharetoa tribal lands, Justice Department land and also Kaimanawa State Forest Park (Fig. 1.2). A major arterial transport route, State Highway 1, runs through the centre of the area as well as three sets of high voltage electricity transmission lines. Parts of the major Tongariro power development are also contained within the area.

2 175°30'

Mt. To ngariro .... Mt. Ngauruhoe ....

0 5 10 km

- - - - Main Trunk railway line

- State Highways (SH)

Figure 1.1. Location of the study area (shaded) within the To ngariro Volcanic Centre, central North Island, New Zealand

3 In this hydroelectric power scheme water is diverted from the southern tributaries of Ruapehu through an underground aqueduct into Lake Moawhango. Subsequently the water travels via a tunnel to the Rangipo Dam within the Tongariro River channel; water from the Wahianoa River is also diverted into the dam (Ministry of Works and Development, 197 4 ). Water is then taken from the dam into the underground Rangipo Power station before being returned to the Tongariro River. A short distance downstream a variable proportion of the water is diverted through the Poutu Tunnel and Canal into Lake Rotoaira. A further tunnel is used to transfer water from Lake Rotoaira to the Tokaanu Power Station and eventually Lake Taupo via the Tokaanu Tailrace Canal (Fig. 1.2). This power development produces not only 1182 GWh of electricity from the two stations per annum, but adds extra water to Lake Taupo which in turn increases power generation at eight further hydroelectric power stations along the Waikato River. At the time of writing, the Rangipo Power Station is generating at only half of its capacity, because one of its two turbines needs to be replaced due to abrasion suffered from sediment in the water sourced from tephra and lahars from Ruapehu eruptions during September-November 1995.

1.5 Climate, soils and land use

The area studied ranges widely in physiography and altitude from the steep slopes of Ruapehu and Tongariro Volcanoes at ea. 1400 m to the flat surfaces of Turangi and surrounds at 360 m. There is a corresponding range in climatic conditions within this area. Turangi has a mean annual temperature of 12 oc with a relatively large range between the average daily maximum and minimum temperatures of 17-6.9 oc {Thompson, 1984). Away from the influence of the ocean, Turangi tends to have warm summer days and very cool winter evenings. The incidence of both ground and air frosts is high and the area is well sheltered from the prevailing westerly wind. Mean annual rainfall is in the order of 1600 mm (N.Z. Meteorological Service, 1980) with maximum falls in the late autumn and winter months and lower falls in the summe�months. This produces a summer soil moisture deficit, while in winter runoff occurs (Tho�pson, 1984 ). In the higher altitude areas of Ruapehu and Tongariro rainfall is on average higher, up to 5000 mm, gale-force winds are common and temperatures are much lower. At 1500 m altitude the mean annual maximum to minimum temperature range is 9.4 to 2.3 oc. Heavy snow falls are common in the alpine areas and the winter snowline descends to the 1400-1500 m level. Four main groups of soils occur within the study area (Fig. 1.3) all of which are azonal and intrazonal soils (Staffof Soil Bureau, 1968; Water and Soils Division, Ministry of Works and Development, 1979). In the areas close to the volcanoes the soils are typically recent soils developed from andesitic tephra erupted from Ruapehu and Tongariro volcanoes since the Taupo Tephra (1850 years B.P.; Froggatt and Lowe, 1990). These soils are known as the Ngauruhoe series and have limited production potential due to the harsh climatic conditions in which they occur.

4 ...... High voltage lines

- State Highways To kaanu series (organic soils) --- Tunnels (for water) recent soils from alluvium Rivers and streams -- Hinemaiaia series (YBPS) = Canals Turangi series (YBPS) - Study area boundaries Rangipo series (YBPS) Ngauruhoe series (recent soils from volcanic ash) [36lbare rock and Ngauruhoe series 39°

Mt. To ngariro A A Mt. Ngauruhoe A A

Ngati Tuwharetoa land 1 I I from 0 5km I Lake Moawhango I I

Figure 1.2. Infrastructure and land ownership within the studied area. Figure 1.3. Soils of the study area. Data from; Staffof Soil Bureau ( 1968) and Water and Soils Division, M.W. and Development (1979). Higher on the flanks of the volcanoes where the degree of erosion is higher areas of bare rock and scree occur. Northeast of the two volcanoes the andesitic tephra covering on top of the Taupo lgnimbrite becomes thinner and the soils are yellow-brown pumice soils, developed directly into the Taupo lgnimbrite. Closer to the volcanoes the soils are Rangipo soils and further north and lower in altitude Turangi soils occur. Both of these soil series have potential for pastoral or exotic forestry use, while the Turangi series with more favourable climatic conditions can also be used for fodder cropping. Within the low-lying area of the Tongariro River delta there is a mixture of recent soils from alluvium in areas near the present river course and organic soils in the swampy areas further away. The alluvium is composed in some cases of reworked Taupo Tephra and gives rise to another type of weakly developed yellow-brown pumice soil, the Hinemaiaia series. These soils are of low fe rtility and have few potential land uses, aside from low producing pastures. The whole area between Turangi and Lake Taupo is low lying and swampy and much of it is covered in peat in which is developed the Tokaanu series, an organic soil. This area has little potential agricultural use as there is difficulty in its drainage and very low gradient to the Lake, so it is best left as conservation reserve. The actual land use in this area depends not only on the soils but also the land ownership or status. Much of the study area, particularly the southern part is undeveloped agriculturally (Fig. 1.4). Portions are within Tongariro National Park and Kaimanawa State Forest Park, and in addition parts of the Ngati Tuwharetoa land is undeveloped. In the northern part of the study area there are large areas of exotic forestry planting (Radiata Pine and Douglas Fir) on the NW slopes of Tongariro and also alongside the Tongariro River. Well developed pastoral lands are restricted to the Rangipo Prison farm and surrounds, and smaller areas near Turangi. Much of the swampy low-lying area between Lake Taupo and Turangi and Tokaanu is either poorly developed pasture or remains as undeveloped wetlands with a high conservation value.

1.6 Regional geologic setting

The current geological setting of New Zealand is dominated by the boundary between the lndo-Australian and Pacific plates which occurs through most of the length of the country (Fig. 1.5). The oblique angle of convergence of the two plates, the type of the crust along the boundary and the geometry of the plate boundary through New Zealand causes variations in the nature of the boundary (Cole and Lewis, 1981; Cole, 1986; Kamp, 1992). To the NE of New Zealand oceanic crust of the Pacific Plate is being subducted beneath the Australian Plate, forming the Tonga-Kermadec Island arc system which includes the Kermadec volcanoes. Farther south the angle of convergence between the two plates and subduction becomes more oblique through the central North Island of New Zealand.

6 I I 175 ° I I Taupo Volcanic I D Zone I I Urban areas Swamp vegetation Plate Boundary llJ ,' �:. rj� - I �I Pasture Major Fault lines .-B> ItJ:: fi Tokaanu � -- £EI� 0 I I c: Exotic Forest •�>9 .fl� lE ! Native or reverting to native .:jl I };:. Subduction in the "§I;! ! tussock, shrub and forest lands I I c..; 0 direction of the (j I � -8 I I 117 arrows II I g <:J---60 � I I � mmty ,, ,..., r

lndo-Australian 50 , Plate <1-= , , mm/yr , 0° o'>OJ� ' 4 ' ' I . "'� I I .s-tti�� I -....! I Mt. To ngariro I I ' � ' .. , ' Bare: 40 .6.rock areas �1"1'1'1{ Mt. Ngauruhoe

Pacific Plate

llJ . I I 5 �I I llJOl..c: 0 llJ I f) c: ll: -�1 >-eIll ·;;:I .,_Q 150 km !J ll:r � 0 er� I ::ll �'lt I I er' .m r � �ti rj r I I 0 5 km I I Figure 1.5. Major elements of the Pacific/lndo-Australian Plate boundarythrough New Zealand, and the positions of the Taupo Volcanic Zone and the Tongariro Figure 1.4. Vegatationllanduse within study area. Volcanic Centre. Adapted from Cole and Lewis (1981). Subduction is expressed in the central North Island by the development of a frontal ridge in E North Island; a volcanic front, including the andesitic and dacitic volcanoes of Ruapehu, Tongariro, Tauhara, Edgecumbe and White Island; and an ensialic marginal basin behind the volcanic front (Cole, 1990). Crustal spreading in the marginal basin has produced the voluminous Taupe Volcanic Zone (TVZ) rhyolitic volcanism in the Central North Island throughout the Quaternary (Wilson, et al. , 1984; Wilson et al., 1986). TVZ volcanism appears to have begun at around 1.6 Ma (Pringle et al., 1992; Soengkono et al., 1992), reflecting the inception of the current orientation of the plate boundary. The southern limit of subduction volcanism is marked approximately by Ruapehu volcano; from this latitude southward the decreasing angle of convergence of the two plates eventually gives rise to largely strike-slip motion between the two plates in the South Island (Fig. 1.5). Strike slip motion is taken up along a series of faults in Hawkes Bay and Wairarapa in the North Island and in the Hope Fault zone and Alpine Fault in the South Island. The component of convergence between the two plates in the South Island is represented by the uplift of the Southern Alps along the Alpine Fault Zone where there is continental-continental crust collision (Kamp et al., 1989; Kamp and Tippett, 1993). In the general vicinity of the study area there are three geologically distinct groups of rocks (Gregg, 1960). These are: (1 ) indurated Triassic-Jurassic greywacke/argillite rocks and their low grade (semi-schist) metamorphic equivalents; (2) weakly indurated Tertiary (Miocene-Piiocene) marine sediments; and (3) Quaternary volcanic rocks and deposits with associated volcaniclastic sediments. The Triassic-Jurassic greywacke and argillite rocks make up the Kaimanawa Mountains in the eastern part of the area and are the basement of the volcanic terrain (Fig. 1.6). These rocks are equivalent to the Torlesse Terrane, an important and widely distributed basement terrane throughout New Zealand (Korsch and Wellman, 1988). These highly indurated and relatively unfossiliferous sedimentary rocks were originally deposited by turbidity currents in a deep ancestral basin and were sourced from a granitic continental mass. The sediments were then scraped from a subducting crustal slab to be accumulated as an accretionary wedge. Later deformation and metamorphism occurred in the late Mesozoic when the rocks were metamorphosed up to semi-schist grade in places. They have been uplifted to their present position over the last 2 million years during the Kaikoura Orogeny, relating to the propagation of the present boundary between the Pacificand lndo-Australian Plates through New Zealand. Tertiary marine sediments within the area (Fig 1.6) were deposited on top of an eroded surface of the Mesozoic greywacke/argillite basement (Fieming, 1953; Gregg, 1960; Grindley, 1960; Williams, 1994). These sediments were deposited in the Miocene and Pliocene and include fossiliferous sandstone and siltstone with lenses of sub bituminous coal, calcic sandstone and limestone units. They represent the encroachment and presence of a shallow sea over the region during Miocene and Pliocene times.

8 Recent alluvium

Pumice alluvium of Taupe Tephra

Recent lahars of Whangaehu River

Waimarino Lahars

Murimotu Lahars late Huatapu lahars Pleistocene Rangipo Lahars l Kawakawa Tephra

Ruapehu Andesite

Tongariro Andesite Quaternary Pihanga Andesite ] Kakaramea Andesite

Pliocene marine sediments

Miocene marine sediments

Triassic greywake/argillite

Triassic Kaimanawa Schist

Main Trunk railway line State Highways (SH)

Figure 1.6. Sketch map of the geology of the region surrounding the study area (adapted from Grindley (1960) and Hay (1967).

9 Quaternary volcanic rocks cap the eroded remnants of these late Tertiary sediments (Fig. 1.6) and mostly consist of a series of andesitic stratovolcanoes of the Tongariro Volcanic Centre which are surrounded by aprons of volcaniclastic sediment and tephras. In addition, rhyolitic ignimbrites and reworked ignimbrite material from the Taupe volcano cover the northern part of the region (Fig. 1.6).

1.7 Tongariro Volcanic Centre

The Tongariro Volcanic Centre (TgVC) comprises four main volcanoes which lie at the southern terminus of the Taupe Volcanic Zone (Fig. 1.5 and 1.6; Plates 1 and 2). In addition to the four large andesitic volcanoes of Ruapehu, Tongariro, Kakaramea-Tihia and Pihanga (in order of descending size) there are several smaller outlying centres. The eruptives of TgVC are mostly calcalkaline medium-K basic and acid andesites with minor amounts of basalt and dacite (Cole et al., 1986; Graham and Hackett, 1987). Within the porphyritic lavas the most common phenocrysts are plagioclase, hypersthene and augite, with minor amounts of hornblende and olivine. The TgVC eruptives are typical erogenic andesites consistent with the upper surface of the subducted Pacific Plate slab being 80 km below (Gamble et al., 1990). The oldest radiometrically dated lava from the TgVC is 0.27 ± 0.02 Ma (Hobden et al., 1996); however, pebbles of TgVC lithology have been found in Lower Pleistocene conglomerates in the Wanganui region to the south (Fieming, 1953). Activity continues into the historic record for both Tongariro and Ruapehu volcanoes with the most recent eruptions being those of Ruapehu volcano in 1995-1996. The pre-50 ka eruptive centres in the TgVC are aligned in a NW orientation, while the post-50 ka centres trend NNE, cross-cutting the old NW alignment (Hackett, 1985; Cole et al., 1986). The eruptives from these two apparent sets of volcanism are petrologically distinct; the younger rocks are lower silica andesites than the older series and they contain more olivine-bearing lavas. The NNE trending vents are aligned with the young andesitic volcanoes of Mt. Edgecumbe, Whale Island and White Island and the orientation of the present subduction system, whereas the older NW trending system may represent an earlier orientation of subduction.

1.8 Previous work in and around the study area

The first mapping of the Tongariro region was carried out by Grange et al., (1938). This reconnaissance mapping outlined the main components of the regional geology and the main volcanic features. Gregg (1960) described early geologic observations and records of the volcanic activity and provided a description of the volcanic and non-volcanic geology of the area including the Grange et al. (1938) maps. Later work by Grindley (1960) distinguished in greater detail several lava and lahar formations on and around Ruapehu. The Ruapehu ring plain was mapped as dominantly Waimarino Lahars (Fig. 1.6) determined to be of Last Glacial age.

10 Plate 1.1: View of the eastern flanks of Ruapehu volcano across its ring plain.

Plate 1.2 View of Tongariro volcano from the south-east. The large cone on the left is the most recently active vent, Ngauruhoe.

11 Two further lahar formations, the Hautapu, and Murimotu Lahars which form terraces in the Whangaehu valley were correlated to stadials of the Last Glacial and recent lahars were mapped in the Whangaehu Valley. Hay (1967) also mapped Murimotu Lahars on the NW flanks of Ruapehu. This deposit has been subsequently described as a 9.5 ka debris-avalanche deposit (Palmer and Neall, 1989). Mathews (1967) mapped and described the Tongariro massif as being a multiple volcano, made up of several small overlapping volcanic cones of varying age. Mathews also mapped paired Last Glacial moraines beside the Waihohonu Valley as well as in two valleys on the eastern flanks of the volcano. The first detailed work on the stratigraphy of the Tongariro and Ruapehu ring plains was carried out by Topping (1974). Topping established a detailed chronology of late Pleistocene and Holocene distal rhyolitic tephras erupted from the TVZ calderas to the north which were interbedded within the andesitic ring plain sequence (Topping and Kohn, 1973). Using these tephra layers as time planes Topping developed a stratigraphy of andesitic tephras erupted from Tongariro and Ruapehu volcano from 13.8 ka B.P. to the present (Table 1.1) and mapped their distributions (Topping, 1973). The Tongariro Subgroup was defined to include all of these described tephra formations. Topping also noted the presence of Hinuera Formation gravels containing reworked Kawakawa Tephra and ranging in age from 22.59 ka B.P. to ea. 10 ka. Hackett (1985) carried out a detailed study of Ruapehu and described four formations of lavas and associated volcaniclastic rocks which comprise the massif (Fig 1.7). Each of these formations represents a cone-building episode of Ruapehu. The Te Herenga Formation is the remnants of the oldest cone-building episode, the vents for which are thought to lie in the N-NW of Ruapehu (Hackett, 1985; Hackett and Houghton, 1989). The source vents for the next youngest, Wahianoa Formation are thought to lie in the SE of the present cone. Both the Te Herenga and Wahianoa Formations are of andesitic composition. The Mangawhero Formation occupies the present summit of Ruapehu and is of basaltic to dacitic composition. The lavas erupted in the post-glacial period belong to the Whakapapa Formation which includes eruptives from the present summit vents as well as several flank vents spread widely over the volcano. The lavas of Ruapehu and the other volcanoes of the TgVC were further subdivided on the basis of petrology and geochemistry by Cole et al. (1986) and Graham and Hackett (1987). They defined six lava types on the volcano and its related vents characterised by their phenocryst species and geochemical similarities. Prior to the present study the most recent work on the eastern ring plains was that of Donoghue (1991 ) who studied in detail the lahar and tephra stratigraphy of the Ruapehu ring plain to the south of the area investigated in this study. Purves (1990), in a landscape ecology study on the Rangipo Desert area of the SE Ruapehu ring plain also contributed to elucidating the lahar stratigraphy of the area. Donoghue ( 1991) and Purves ( 1990) significantly revised the lahar stratigraphy from the early mapping of Grindley (1960) and defined five new lahar formations dating from > 22.6 ka B.P to the present (Table 1.2).

12 Table 1.1. Stratigraphy of Tongariro Subgroup tephras and interbedded rhyolitic tephras described by Topping (1973, 1974) and Topping and Kohn (1973). Tephra ages are those currently accepted.

Formation* Members Source Age ka Reference to age t Ngauruhoe Tephra TgVC ea. 1.8-01 Taupo Tephra (Unit Y) Taupo 1.852 Froggatt and Lowe (1990) Mangatawai Tephra Tongariro 2.5-1.852 unnamed andesitic ash Whakaipo Tephra Taupo unnamed andesitic ash TgVC Waimahia Tephra (Unit S) Taupo Papakai Tephra TgVC Hinemaiaia Tephra Taupo 5.2-3.952 Wilson (1993) Papakai Tephra TgVC Rotoma Ash Oakataina 8.532 Froggatt and Lowe (1990) Papakai Tephra TgVC Opepe Tephra (Unit E) Taupo 9.052 Papakai Tephra TgVC Mangamate Tephra Poutu Lapilli Tongariro ea. 9.71 Topping (1973) Wharepu ea. 9.71 Tephra Poronui Tephra (Unit C) Taupo 9.812 Froggatt and Lowe (1990) Mangamate Tephra Ohinepango Tongariro ea. 9.71 Topping (1973) Tephra Waihohonu ea. 9.71 Lapilli Oturere Lapilli 9.782 Karapiti Tephra (Unit B) Taupo 10.12 Wilson (1993) Okupata Tephra Ruapehu >9.792 Topping (1973) unnamed andesitic ash TgVC ?Rotorua Tephra Okataina 13.082 Froggatt and Lowe (1990) unnamed andesitic ash TgVC Puketarata Ash Maroa unnamed andesitic ash TgVC Rotoaira Tephra Tongariro 13.82 Topping (1973) unnamed andesitic ash TgVC Rerewhakaaitu Tephra Okataina 14.72 Froggatt and Lowe (1990) unnamed andesitic ash Tongariro Kawakawa Tephra Oruanui Taupo Wilson et al., (1988) lgnimbrite and Aokautere Ash * Taupo Volcanic Centre Unit nomenclature from Wilson (1993). TgVC = Tongariro and Ruapehu volcanoes, Tongariro includes the vent of Ngauruhoe. t 1 Estimated ages. 2 Average or single 14C ages on old (Libby) half life basis.

13 Mt. Ngauruhoe � National Park

� Vents less than or equal to 30 ka 0 Skm

Ruapehu Group N.Z. Stages K-Ar ages (Stipp, 1968)

Whakapapa Formation Aranuian

---- ·unconformity ----- .1-===-=-=i ------24 ka Mangawhero Formation Otiran 36 ka

Wahianoa Formation Oturian

230 ka Te Herenga Formation Terangian

Figure 1.7. Generalised geologic map of Ruapehu volcano and related vents, adapted from Hackett (1985), with additions from Donoghue et a/ (1988) and Lecointre et a/ (1994).

14 Table 1.2. Stratigraphy of lahar fo rmations on the SE Ruapehu ring plain, from Donoghue (1991) and Purves (1990).

Lahar formation Age range (ka B.P.) Tephra boundary marker beds Onetapu Formation 1 .85 - present Taupo Tephra Waimahia Tephra Manutahi Formation 3.28 - 4.60 Mangaio Formation 1 4.60 Manutahi Formation 4.60 - 5.43 Motutere Tephra Tangatu Formation 5.43 - 14.7 Rerewhakaaitu Tephra Te Heuheu Formation 14.7 - > 22.6 no lower limit defined 1 Mangaio Formation represents the deposits of a single event or a closely spaced series of events dated at 4.6 ka B.P. by Donoghue (1991). The other formations span the presented time ranges and include multiple events.

Based on the defined and mapped lahar fo rmations Donoghue (1991) and Hodgson (1993) produced lahar hazard maps for the SE Ruapehu ring plain and the Whangaehu catchment. Do nog hue (1991) and Donoghue et al. (1995) also added to and revised the previously established rhyolitic and andesitic tephrostratigraphy developed by Topping (1973, 1974) and Topping and Kohn (1973) extending the stratigraphy to 22.6 ka B.P. Donoghue et al. (1995) revised the base of the Tongariro Subgroup from ea. 14 ka B.P. to ea. 10 ka B.P. and defined the Tukino Subgroup to include tephras aged between 10 and 22.6 ka B.P. They group all tephras erupted from the Tongariro Volcanic Centre since 22.6 ka B.P. into seven formations (Table 1.3). The major additions to the existing stratigraphy were the newly defined Tufa Trig and Bullet Formations. The Tufa Trig Formation with 18 members and the Bullet Formation with 23 lapilli members record the activity of Ruapehu volcano over the last 1.85 ka B.P. and the period 10 to 22.6 ka B.P. respectively. The definition of these two formations and mapping of the tephras within them contributed significantlyto knowledge of the eruptive history of Ruapehu volcano.

1.9 Summary and conclusions from previous work

Geologic mapping of the study area and the surrounding region until the 1970's was concentrated on the volcanic cones with only reconnaissance work carried out on the ring plains. The ring plains of Ruapehu and Tongariro volcanoes were mapped as mostly Waimarino and Rangipo Lahars respectively. The Ruapehu ring plain also had smaller areas mapped in more detail. Topping (1973 and 1974) and Topping and Kohn (1974) constructed the first detailed late Pleistocene and Holocene andesitic and rhyolitic tephrostratigraphy of the area, defining the Tongariro Subgroup.

15 Table 1.3. Stratigraphy of andesitic and rhyolitic tephras on the SE Ruapehu ring plain, from Donoghue (1991 ) and Donoghue et al. (1995).

Subgroup Formation Members Source Age (ka) Interbedded rhyolitic te hras

Ngauruhoe Formation Tongariro ea. 1.85 -01 Kaharoa Tephra Tufa Trig Formation Tf1 8 - Tf1 Ruapehu ea. 1.85 -01 Kaharoa Tephra 1.852 Taupo Tephra Mangatawai Tephra Tongariro 2.5 - 1.852 Mapara Tephra Papakai Formation TgVC 2.5 - 9.71 Waimahia Tephra Hinemaiaia Tephra Whakatane Tephra Motutere Tephra Tongariro Mangamate Poutu Lapilli Tongariro ea. 9.71 Subgroup Formation Wharepu Tephra 9.812 Poronui Tephra Ohinepango Tephra Waihohonu Lapilli Oturere Lapilli T e Rato Lapilli 9.782 unnamed tephra TgVC 10Y Karapiti Tephra unnamed tephra TgVC Pahoka T ephra Tongariro ea. 10-9.81 Bullot Formation Ruapehu ea. 101 (upper) " Ngamatea Lapilli-2 Ngamatea Lapilli-1 Pourahu Member L18 -L17 Tu ino Shawcroft Subgroup Lapilli L16 11.852 Waiohau Tephra L 15 - L8 13.082 Rotorua T ephra ? Rotoaira Lapilli Tongariro 13.82 Bullot Formation L 15- L8 Ruapehu (upper) 14.72 Rerewhakaaitu Tephra Bullot Formation L7b - L4 (middle) ea. 18.01 Okareka Tephra3 Bullot Formation L3 - L 1 (lower) 22.62 Kawakawa Tephra

1 Estimated ages, references to ages the same as Table 1.1. 2 Average or single 14C ages on old (Libby) half life basis, references as for Table 1.1. 3 Age from Nairn (1992).

16 Hackett (1985) described the stratigraphy of the Ruapehu cone and with Cole et al. (1986) and Graham and Hackett (1987) subdivided the lavas of TgVC on the basis of their mineralogy and geochemistry. Later work by Donoghue (1991) and Purves (1990) constructed a more detailed lahar stratigraphy of the SE Ruapehu ring plain with 5 lahar formations defined by dated tephra interbeds. Donoghue (1991) and Hodgson (1993) then were able to produce lahar hazard maps of the SE Ruapehu ring plain and Whangaehu River based on the lahar stratigraphy. Donoghue (1991) and Donoghue et al. (1995) revised and added to the existing tephrostratigraphy, adding the Tukino Subgroup to extend the stratigraphy to 22.6 ka B.P. Donoghue and eo-workers also added a new stratigraphy of Ruapehu-sourced eruptives on the SE Ruapehu ring plain. In conclusion, there are fo ur areas in which geological investigations in the study area can add to that which has previously been carried out: (1) The NE Ruapehu lahar and tephra stratigraphy up to 22.5 ka B.P. can be mapped using the framework established on the SE Ruapehu ring plain. (2) The ring plain record older than 22.5 ka B.P. can be investigated to extend the existing lahar and tephra stratigraphy and the eruptive history of Ruapehu volcano. Both of these first two areas of investigation can contribute to a synthesis of the construction of the NE Ruapehu ring plain. (3) The E Tongariro ring plain has not been investigated since the reconnaissance mapping of Grindley (1960) in which the entire area was mapped as the Rangipo Lahars. An investigation of the lahar and tephra stratigraphy in this area can potentially provide further information on the eruptive history of Tongariro volcano. (4) By mapping the lahar stratigraphy on the NE Ruapehu and E Tongariro ring plains as well as along the Tongariro River a lahar hazard map for these areas can be constructed. In particular, knowledge of the lahar history of the Tongariro River, one of the most important drainage channels from TgVC is necessary because the town of Turangi (ea. 4200 population) is located on low surfaces beside the river. In addition the river is used for the Tongariro Power Development and all of its tributaries draining the volcanoes are crossed by the main transport arterial route of State Highway 1. The relevance of this issue is demonstrated by the generation of at least two lahars in the Mangatoetoenui Stream in 1995 which entered the Tongariro River. These lahars have seriously impacted on the Tongariro Power development by causing abrasion of turbines in the Rangipo Power Station necessitating the replacement of one of the two turbines, and also debris has accumulated behind the Rangipo Dam severely reducing its storage capacity. Serious effects on trout (a major tourist attraction and source of revenue for Turangi) also occurred and continue to occur at the time of writing due to much additional fine sediment remaining in the river system.

17 1.1 0 References

Clarke, C. and Smith, M., 1986. Turangi town centre, where to now? Taupe County Council Special Report, 15. Cole, J.W., 1986. Distribution and tectonic setting of late Cenozoic volcanism in New Zealand. Royal Soc. of N.Z. Bull., 23: 7-20. Cole, J.W., 1990. Structural control and origin of volcanism in the Taupe Volcanic Zone, New Zealand. Bull. Volcanol., 52: 445-492. Cole, J.W. and Lewis, K.B., 1981. Evolution of the Taupo-Hikurangi subduction system. Tectonophysics, 72; 1-21. Cole, J.W., Graham, I.J., Hackett, W.R., Houghton, B.F. , 1986. Volcanology and petrology of the Quaternary composite volcanoes of Tongariro Volcanic Centre, Taupe Volcanic Zone. Royal Soc. of N.Z. Bull., 23: 224-250. Department of Statistics, 1992. 1991 Census of Population and Dwellings, Waikato/Bay of Plenty Regional Report. Department of Statistics, Wellington, N.Z. Donoghue, S.L., Neall, V.E., Palmer, A.S., 1988. Holocene geology of the upper Whangaehu River, Mt. Ruapehu. Geological Soc. of N.Z. Mise. Publ. 41 a: 61. Donoghue, S.L., 1991 . Late Quaternary volcanic stratigraphy of the south-eastern sector of Mount Ruapehu ring plain, New Zealand. Unpub. PhD. thesis, Massey University, Palmerston North, N. Z. Donoghue, S.L., Neall, V.E., Palmer, A.S., 1995. Stratigraphy and chronology of late Quaternary andesitic tephra deposits, Tongariro Volcanic Centre, New Zealand. J. Ray. Soc. of N.Z., 25: 115-206. Fleming, C.A., 1953. The Geology of Wanganui Subdivision. N.Z. Geological Survey Bull., 52: 362 pp. Froggatt, P.C. and Lowe, D. J., 1990. A review of late Quaternary silicic and some other tephra formations from New Zealand: their stratigraphy, nomenclature, distribution, volume and age. N.Z. J. Geol. and Geophys., 33: 89-109. Gamble, J.A., Smith, I.E.M., Graham, I.J., Kokelaar, B.P., Cole, J.W., Houghton, B.F., Wilson, C.J.N., 1990. The petrology, phase relations and tectonic setting of basalts from the Taupe Volcanic Zone, New Zealand, and Kermadec Island Arc Havre Trough, SW Pacific. J. Volcano!. and Geotherm. Res., 54: 265-290. Graham, I.J., Hackett, W.R., 1987. Petrology of calc-alkaline lavas from Ruapehu volcano and related vents, Taupe Volcanic Zone, New Zealand. J. of Petrology, 28: 531-567. Grange, L.l., Williamson, J.H., Hurst, J.A., 1938. Geological maps of Tongariro Subdivision, to accompany, N.Z. Geological Survey Bull, 40: 4 sheets. Gregg, D.R., 1960. The geology of Tongariro Subdivision. N.Z. Geological Survey Bull., 40: 152pp. Grindley, G.W., 1960. Geological map of New Zealand, 1st ed., 1: 250 000, Sheet 8, Taupe. D.S.I.R., Wellington.

18 Hackett, W.R., 1985. Geology and petrology of Ruapehu volcano and related vents. Unpub. thesis Victoria University of Wellington: 312 pp. Hackett, W.R., Houghton, B.F., 1989. A facies model for a Quaternary andesitic composite volcano: Ruapehu, New Zealand. Bull. Volcano!., 51: 51-68. Hay, R,F., 1967. Geological map of New Zealand, 1st ed., 1: 250 000, Sheet 7, Taranaki. D.S.I.R., Wellington. Hobden, B.J., Houghton, B. F., Lanphere, M.A., Nairn, I.A., 1996. Growth of the Tongariro volcanic complex: new evidence from K-Ar age determinations. N.Z. J. Geol. and Geophys., 39: 151-154. Hodgson, K.A., 1993. Late Quaternary lahars from Mount Ruapehu in the Whangaehu River valley. Unpub. PhD. thesis, Massey University, Palmerston North, N. Z. Kamp, P.J.J., 1992. Tectonic architecture of New Zealand. In Soons, J.M. and Selby, M.J. (eds). Landforms of New Zealand, 2nd Edition, Longman Paul, Auckland: 1- 30. Kamp, P.J.J., Green. P.F., White, S.H., 1989. Fission track analysis reveals character of collisional tectonics in New Zealand. Tectonics, 8: 169-195. Kamp, P.J.J, Tippett, J.M., 1993. Dynamics of Pacific Plate crust in the South Island (New Zealand) zone of oblique continent-continent convergence. J. Geophys. Res., 98(B9): 16 105-16 118. Korsch, R.J., Wellman, H.W., 1988. The geological evolution of New Zealand and the New Zealand region. In Nairn, A.E.M., Stehli, F.G., Uyeda, S. (eds). The Ocean Basins and margins, Vol. 7B: 41 1-481 . Plenium Publishing Corporation. Lecointre, J.A., Neall, V.E., Palmer, A.S., 1994. A new lava field from the southwestern quadrant of the Ruapehu ring plain. Geological Soc. of N.Z. Mise. Publ. BOA: 111. Mathews, W.H., 1967. A contribution to the geology of the Mount Tongariro massif, North Island, New Zealand. N.Z. J. Geol. and Geophys., 10: 1027-1038. Mercer, E.V., (ed), 1973. Turangi "the town of the future". Papers from the Lions symposium 12-13 May 1973. Turangi Lions Club Inc., Turangi. Ministry of Works and Development, 1974. New Zealand's latest power story, Tongariro Power Development. Ministry of Works and Development, Wellington, N.Z. Nairn, I.A., 1992. The Te Rere and Okareka eruptive episodes - Okataina Volcanic Centre, Taupo Volcanic Zone, New Zealand. N.Z. J. Geol. and Geophys., 35: 93- 108. New Zealand Meteorological Service, 1980. Summary of climatological observations to 1980. New Zealand Meteorological Service Mise. Publ., 177. Palmer, B.A., Neall, V.E., 1989. The Murimotu Formation - 9500 year old deposits of a debris avalanche and associated lahars, Mount Ruapehu, North Island, New Zealand. N.Z. J. Geol. and Geophys., 32: 477-486. Pringle, M.S., McWilliams, M., Houghton, B.F., Lanphere, M.A., Wilson, C.J.N., 1992. 40Arf9Ar dating of Quaternary feldspar: Examples from the Taupo Volcanic Zone, New Zealand. Geology, 20: 531-534.

19 Purves, A.M., 1990. Landscape ecology of the Rangipo Desert. Unpub. MSc. Thesis. Massey University, Palmerston North, N.Z. Soengkono, S., Hochstein, M.P., Smith, I.E.M., ltaya, T., 1992. Geophysical evidence for widespread reversely magnetised pyroclastics in the western Taupe Volcanic Zone (New Zealand). N.Z. J. of Geol. and Geophys., 35: 47-55. Staffof Soil Bureau, 1968. Soils of New Zealand, Part 1. N.Z. Soil Bureau Bull. 26 (1 ). Stipp, J.J., 1968. The geochronology and petrogenesis of the Cenozoic volcanics of the North Island of New Zealand. Unpub. Ph.D. Thesis. Australian National University, Canberra. Thompson, C.S., 1984. The weather and climate of the Tongariro Region. N. Z. Meteorological Service Mise. Publ., 115(14): 35pp. Topping, W.W., 1973. Tephrostratigraphy and chronology of late Quaternary eruptives from the Tongariro Volcanic Centre, New Zealand. N. Z. J. Geol. and Geophys., 16: 397-423. Topping, W.W., 1974. Some aspects of Quaternary history of Tongariro Volcanic Centre. Unpub. PhD. thesis, Victoria University of Wellington N. Z. Topping, W.W., Kohn, B.P., 1973. Rhyolitic tephra marker beds in the Tongariro area, North Island, New Zealand. N.Z. J. Geol. and Geophys., 16: 375-395. Water and Soils Division, Ministry of Works and Development, 1979. N. Z. Land Resource Inventory Worksheets, N102 Tokaanu and N112 Ngauruhoe and Bay of Plenty - Volcanic Region Landuse Capability Extended Legend (12pp). National Water and Soil Conservation Organisation, Ministry of Works and Development, Wellington, N.Z. Williams, J.K., 1994. Some aspects of the Cenozoic geology of the Moawhango River region, in the army training area, Waiouru, North Island, New Zealand. Unpub. MSc. Thesis. Massey University, Palmerston North, N.Z. Wilson, C.J.N., Rogan, A.M., Smith, I.E.M., Northey, D.J., Nairn, I.A., Houghton, B.F., 1984. Caldera volcanoes of the Taupe Volcanic Zone, New Zealand. J. Geophys. Res., 89(B10): 8463-8484. Wilson, C.J.N, Houghton, B.F., Lloyd, E.F., 1986. Volcanic history and evolution of the Maroa-Taupo area, central North Island. Royal Soc. of N.Z. Bull., 23: 194-223. Wilson, C.J.N., Switzur, R.V., Ward, A.P., 1988. A new 14C age for the Oruanui (Wairakei) eruption, New Zealand. Geol. Mag., 125: 297-300. Wilson, C.J.N., 1993. Stratigraphy, chronology, styles and dynamics of late Quaternary eruptions from Taupe Volcano, New Zealand. Phil. Trans. Roy. Soc. London A 343: 205-306.

20 CHAPTER 2: RHYOLITIC TEPHROCHRONOLOGY

2.1 Introduction

Rhyolitic tephra layers have played a key role in previous stratigraphic studies of the Tongariro and Ruapehu ring plains. In particular, Topping and Kohn (1973) and Donoghue et al. (1995) demonstrated the use of dated rhyolitic tephra layers to establish a stratigraphy of andesitic tephra layers and lahar deposits. In this study, one of the first approaches to establishing a stratigraphy on the NE Ruapehu and E Tongariro ring plains was to find and identify rhyolitic tephras in described sequences. In situations where the stratigraphy of a site was otherwise unclear, identification of a rhyolitic tephra layer within the sequence was an invaluable method of dating the sequence. In addition, in the older part of the ring plain sequences (> 40 ka B.P.), radiocarbon dating is impossible and identifying a rhyolitic tephra within a sequence became even more crucial. The major problem encountered with using rhyolitic tephra layers was establishing their identity with confidence. Previous methods of rhyolitic tephra identification on the Tongariro and Ruapehu ring plains were based on stratigraphy, mineralogy and numerical comparisons of glass chemistry (Topping and Kohn, 1973; Donoghue, 1991 ). However, in many instances these methods were found to be insufficient to uniquely identify a particular tephra. Usually the existing methods could only narrow the field of possibilities, resulting in the necessity for a subjective decision of the tephras identity. In this study, methods of improving the unique identification of rhyolitic tephras within the ring plain sequence were investigated. In particular the statistical analysis of major oxide glass chemistry was attempted. These studies form the paper below which is in press in the New Zealand Journal of Geology and Geophysics (Vol. 40 No. 2, 1997). The paper has multiple authors and the following lists each author's contribution to the work:

S. J. Cronin: Principal investigator Carried out all: field description and sampling, laboratory preparation of samples, optical microscopy work, electron microprobe analysis, statistical programming and analysis, manuscript preparation and writing

V.E. Neall, A.S. Palmer, R.B. Stewart:Adviser s Aided the study by: discussing results and methodology with SJC editing and discussion of the manuscript

21 2.2 Additional methodology and notes

The problem of statistical closure discussed in the manuscript is due to the nature of compositional data; the component proportions (in this case, oxide weight proportions) must sum to unity (or close to unity) in compositional data. This is termed the constant sum constraint and causes a high degree of correlation or interdependence between the individual variables making up a composition (Aitchison, 1983 and 1986). However, discriminant function analysis (DFA) requires variables to be independent of one another. To remove the high correlation between the variables the log ratio transformation was developed by Aitchison (1983). The formula:

Yi = log(xi I xd ) describes the transformation, where is the score of the oxide chosen arbitrarily as the xd divisor, some raw oxide score and Y; the transformed score (i can take values 1 to d- 1 ) . X; Not only does this procedure remove the statistical closure it reduces all of the composition variables to the same order of magnitude. The SAS programs used for the DISCRIM, CANDISC and STEPDISC statistical analyses in the following study are presented in Appendix 1. A listing of all electron microprobe analyses of rhyolitic tephra samples carried out in this study comprises Appendix 2.

2.3 METHODS OF IDENTIFYING LATE QUATERNARY RHYOLITIC TEPHRAS ON THE RING PLAINS OF RUAPEHU AND TONGARIRO VOLCANOES, NEW ZEALAND

Shane J. Cronin, V.E. Neall, A.S. Palmer and R.B. Stewart Department of Soil Science, Massey University, Private Bag 11 222, Palmerston North, New Zealand.

New Zealand Journalof Geology and Geophysics 40, No. 2, in press.

Abstract On the ring plains of Ruapehu and Tongariro volcanoes, distal rhyolitic marker tephras provide a valuable stratigraphic framework. However, identification of many of these tephras has been imprecise. Here we provide a quantitative approach for identifying tephras within the ring-plain sequences. We extend from simple canonical discriminant function models of glass chemistry to show how these, in conjunction with other geological information, can be used in a practical field-based study. In two stratigraphically distinguishable groups (1 0-22 ka and 22-65 ka), we established discriminant models for possible tephra correlatives from standard glass analyses. Testing analyses from unknown tephras against the models classified 34 of the 41 samples with probabilities > 0.75 to tephras that were consistent with mineralogical and stratigraphic evidence. Unknowns with lower probabilities of classification had several possible correlatives. Some of these were improved when the tephras classified with > 0.75 probability, and which were consistent with stratigraphic and other evidence, were

22 added to the discriminant models. The improved classifications were due to an increased number of samples for each tephra and also because the added analyses were produced by the same EMP operator under the same instrument conditions. Classifications of other unknowns were improved by considering them as mixed tephras. In addition to more rigorously correlating several tephras previously identified in this area, four tephras have been identified in the area for the first time, the Okaia, Omataroa and Hauparu Tephras and the Rotoehu Ash. These occur as microscopic accumulations of rhyolitic glass shards within weathered andesitic tephra deposits.

2.4 Keywords Ruapehu volcano, Tongariro volcano, ring plain, stratigraphy, rhyolitic tephra correlation, discriminant function analysis, Okaia Tephra, Omataroa Tephra, Hauparu Tephra, Rotoehu Ash.

2.5 Introduction

Tephra layers play an important part in stratigraphic studies throughout New Zealand and the surrounding sea floor. The most voluminous and best studied tephra layers are the rhyolitic eruptives from calderas of the Taupo Volcanic Zone in the central North Island of New Zealand. Over 30 dated, widespread, rhyolitic tephras have erupted over the last c. 65 ka and most have well-constrained radiocarbon ages (Vucetich and Pullar, 1969; 1973; Howorth, 1975; Vucetich and Howorth, 1976; Froggatt and Lowe, 1990). Primary tephra layers enable isochronous time horizons to be established in a wide range of depositional environments. In studies of the ring plains of andesitic volcanoes within the Tongariro Volcanic Centre, rhyolitic tephras enable relative ages to be assigned to andesitic tephras and lahar deposits in order to provide a basis for calculating eruption and volcanic hazard frequencies (Cronin et al., 1996a; Cronin and Neall, in press). The tephras also are used for investigating the construction of the ring plains (Cronin et al., 1996b) and for investigations of soil development on ring plain materials (Cronin et al., 1996c). Methods of rhyolitic tephra identification have become increasingly sophisticated over time as the need arose for identification of tephras in distal areas, where field criteria are less useful. Methods applied have involved mineralogical studies (e.g. Ewart, 1963; Randle et al., 1971; Lowe, 1988) and chemical studies (e.g. Smith and Westgate, 1969; Kahn, 1970; Borchardt et al., 1971 ; Froggatt, 1983; Stokes et al., 1992). Chemical techniques have concentrated on glass and titanomagnetite, and in New Zealand, glass chemistry has assumed the greatest importance for chemical characterisation of tephras (e.g. Froggatt, 1983; Stokes et al., 1992). Stokes and Lowe (1988) introduced the use of canonical discriminant functions (DFA) with glass chemistry to New Zealand tephras and showed a theoretical discrimination of tephras from different volcanic sources. Stokes et al. ( 1992) went on to show a theoretical discrimination of several individual rhyolitic tephras from Okataina and

23 Taupe sources. Shane and Froggatt (1994) again demonstrated a theoretical discrimination of six rhyolitic tephras of widely varying ages from the Taupe Volcanic Zone. All of these studies have shown the potential of the DFA method to discriminate New Zealand rhyolitic tephras but did not go on to show these techniques used in a practical situation. The Shane and Froggatt (1994) study in particular, although providing a further demonstration of the statistical techniques, used tephras which ranged in age from 1.85 ka to 1.63 Ma that are unlikely to be confused stratigraphically. Eighteen distal rhyolitic tephras have been identified on the ring plains of Tongariro and Ruapehu volcanoes ranging in age from ea. 0.8 to 22 ka. (Topping and Kohn, 1973; Donoghue et al. , 1995). In this study we focus on the time range between 10 and 65 ka and provide a practical demonstration of the use of discriminant function analysis (DFA) to identify the rhyolitic tephras present. We extend from the initial construction of DFA models by testing unknown tephras against the models and attempt different means of improving their classification performance.

2.6 Setting

Ruapehu volcano within the Tongariro Volcanic Centre, is a large, active, andesitic composite volcano, and forms the highest point in the North Island at 2797 m (Fig. 2.1 ). 3 The current massif comprises a 110 km cone with a volcaniclastic ring plain of similar volume surrounding it (Hackett and Houghton, 1989). Tongariro volcano is a slightly smaller andesitic massif made up of several coalescing volcanic cones (Mathews, 1967), the largest of which is the recently active cone of Ngauruhoe. The ring plains of Ruapehu and Tongariro volcanoes are confined to the east by the Kaimanawa Mountains, and the Tongariro River is the approximate boundary of the eastern ring plains. The section localities where rhyolitic tephras were found in this study are shown in Fig. 2.1.

2.7 Methods

2. 7.1 Sample Preparation and analysis Rhyolitic tephra samples from macroscopic layers were cleaned in water using an ultrasonic probe; it was unnecessary to concentrate further glass. When tephra samples were scattered within fine andesitic ash or soil material they were cleaned in 0.2 M acid oxalate reagent to disperse them and to remove short-range order and organic constituents (Biakemore et al. , 1987; Alloway et al., 1992). Glass was then concentrated 3 using a sodium polytungstate solution at a density of 2.45 Mg/m . Ferromagnesian minerals were concentrated from all samples using a Frantz lsodynamic Separator. Grain mounts of ferromagnesian minerals were prepared and assemblages described by point counting. Glass grains were mounted in epoxy resin plugs, which were ground to expose the grains and polished.

24 175° 45'

Mt. To ngariro ...

Mt. Ngauruhoe ...

39° 10 I

Wa ihohonu Stream 41

Mt. Ruapehu ...

0 5km - State Highways

Figure 2.1 . Location map of the eastern ring plains of To ngariro and Ruapehu volcanoes. Numbered dots indicate locations of sections where the rhyolitic tephras of this study were sampled; grid references are given in Ta ble 2.2.

25 The major element chemistry was determined for polished individual grains on a JEOL- 733 electron microprobe following the procedures and analytical conditions of Froggatt (1983). A 20 )lm beam diameter was used when possible, otherwise a 10 llm beam was used with a beam current of 8 nA at an accelerating voltage of 15 kV. For each tephra sample at least 10 individual glass shards were analysed. The glass chemistry was compared with published glass analyses carried out under the same analytical conditions and on the same instrument (AIIoway, 1989; Oonoghue, 1991; Froggatt, 1982; 1983; Froggatt and Solloway, 1986; Froggatt and Rogers, 1990; Lowe, 1988; Pillans and Wright, 1992; Stokes et al., 1992; Pillans et al., 1993; Shane and Froggatt, 1994; Carter et al.. 1995). Similarity coefficients, and coefficients of variation (Borchardt et al., 1971 ), were calculated to compare the unknown samples with large volume tephra units in the same general time frame. However, these often provided two or three possible correlatives for a sample, even after their position in the stratigraphic column was considered, leaving the difficulty of choosing the most likely correlative. To improve on this correlation, canonical OFA was used.

2.7.2 Statistical Methods Canonical OFA is a technique related to principal component analysis, which reduces the dimensionality of data such as compositional data which has a large number of independent variables. Canonical OFA produces a small number of linear combinations of the quantitative variables which best discriminate pre-defined groups of observations or analyses. This means that instead of working with ten oxide scores to discriminate groups of samples, one or two canonical variables contain the information. A reference set of analyses with pre-defined groupings must first be set up and canonical OFA is used to produce a discriminant model which can be used to classify unknown observations or analyses. The theory of OFA is outlined in many texts on multivariate analysis (e.g. Srivastava and Carter, 1983; Johnson and Wichern, 1992). The 02 or Mahalanobis distance statistic is produced within the results of canonical OFA, and indicates the multivariate spacing between data groups in multi-dimensions (Srivastava and Carter, 1983). The 02 statistic is therefore a useful measure of the separation of groups of samples (Stokes et al., 1992) in place of the coefficients of variation and similarity coefficients first used with volcanic glass by Borchardt et al. (1971 ). The studies of Beaudoin and King (1986) and Stokes and Lowe (1988), introduced the use of stepwise OFA methods which selected variables that had the greatest discriminating power. Stepwise OFA is further described by Srivastava and Carter (1983). In this study the SAS system programs OISCRIM, STEPOISC and CANOISC were used (SAS Institute Inc., 1989). Prior to statistical analysis the data used in this study were transformed following the log-ratio procedure of Aitchison (1983, 1986) and Stokes and Lowe (1988) to avoid statistical closure, inherent in compositional data. We used MgO as the oxide divisor for this study, because of its moderate content and relatively low within, and between sample variance (Shane and Froggatt, 1994 ).

26 2.7.3 Tephra classification methodology From the overall time interval of 10-65 ka, separate data bases were established for two stratigraphically distinct groups of tephras (Table 2.1 ); those above and below the regional marker horizon of the Kawakawa Tephra (22.6 ka B.P., Wilson et a/., 1988). This enables two simpler discriminant models to be created rather than a single very complex model. All known tephras of significant volume within each time interval were included. Database 1 contained glass analyses of Karapiti, Waiohau, Rotorua, Rerewhakaaitu, Okareka and Te Rere Tephras. Database 2 contained analyses of Kawakawa, Okaia, Tihoi, Omataroa, Mangaone and Hauparu Tephras, and Rotoehu Ash. Analyses of Maketu Tephra were unavailable. The analyses were from published sources listed previously and unpublished data of SJC. Discriminant models were then created for each of these reference sets using the DISCRIM program. Analyses of samples from the two stratigraphic intervals collected from the Ruapehu and Tongariro ring plains were tested against the discriminant functions calculated from the models. To improve any poor classifications resulting from this first model two further approaches were used. Firstly, unknown tephras that were correctly classified were added to the initial model to attempt to improve its performance in classifying the remaining unknowns. To satisfy the 'correctly classified' precondition, the overall probability of classification of a given sample with a particular tephra had to be high (>0.75), and this had to be consistent with all other evidence from stratigraphy, field observations and mineralogy. Secondly, in some cases where it appeared the sample was of a mixed origin, each component of the mixture was treated as a separate tephra for classification. Once again, any classifications made had to be consistent with other lines of evidence.

Table 2.1 Ages and sources of rhyolitic tephras within the two discriminant models used in this study. Tephra name Source* Symbol Age (Y rs B.P.) Reference for age

Database ka tephras Karapiti Tephra1, 10-22 TVC Kp 9820 ± sot Froggatt & Lowe (1990) Waiohau Tephra OVC Wh 11850 ± 60t Rotorua Tephra OVC Rr 13080 ± sot Rerewhakaaitu Tephra OVC Rk 14700 ±110t

Okareka Tephra OVC Ok c. 18000* Te Rere Tephra OVC Te 21 100 ± 320t

Database ka tephras Kawakawa2, Tephra 22.5-65 TVC Kk 22590 ± 230t Wilson et al. (1988)

Okaia T ephra TVC 0 c. 23000* Froggatt & Lowe (1990) Omataroa Tephra OVC Om 28220 ± 630t Mangaone Tephra OVC Mn 27730 ± 350t Hauparu Tephra OVC Hp 35870 ± 1270t

Tihoi Tephra TVC T c. 46000* Rotoehu Ash OVC Re 64000 ± 4000§ Wilson et al. (1992)

* TVC - Taupo Volcanic Centre; OVC = Okataina Volcanic Centre. t Denotes 14C ages on old half life basis.

* Estimated stratigraphic age. s Whole rock K-Ar age.

27 2.8 Results

2.8.1 Ferromagnesian Assemblages

Table 2.2 Ferromagnesian mineral assemblages (modal %) and location of unknown tephra samples on the Ruapehu ring plain. Ferromagnesian mineralogy Sample Section Grid Ref. number number NZMS 260 opx cpx hb bi cu tm

10-22 ka tephras 93.5 20 T20/465127 52 41 5 tr 2 93.6 22 T20/453129 54 40 2 4 93.7 107 T20/430127 49 37 12 2 93.8 26 T20/435129 43 53 2 2 93.9 29 T20/498157 54 25 5 tr tr 15 93.32 7 T20/469100 53 22 3 18 1 3 93.59 68 T19/505248 69 24 tr 7 94.7 20 T20/465127 48 16 15 21 94.9 30 T20/468162 50 30 1 19 94.21 55 T19/48121 1 58 40 tr 2 94.24 57 T1 9/486211 57 29 14 94.27 58 T19/485221 69 27 2 2 94.28 59 T19/495220 41 16 37 6 94.29 59 T19/495220 46 34 4 16 94.37 62 T19/485238 61 35 3 1 94.47 66 T19/496238 53 23 18 6 94.51 67 T19/504237 56 40 3 1 94.58 68 T19/505248 73 20 2 5 94.71 70 T19/488242 53 34 12 tr 1 94.74 71 T19/499247 68 26 2 4 94.79 73 T19/364544 67 15 6 2 95.2 81 T19/430238 61 37 1 1 95.3 81 T19/430238 53 35 8 tr 4 95.6 82 T19/423339 51 41 4 4 95.9 83 T19/448326 63 32 2 3 95.13 96 T19/549328 59 32 1 8

22.5-65 ka tephras 93.1 1 T20/455088 30 44 18 8 93.3 3 T20/465095 53 37 2 8 93.17 41 T20/461 170 35 18 33 14 93.33 7 T20/469100 32 15 37 16 7.37 7 58 39 3 7.36 7 66 28 6 9.37 9 T20/468102 63 27 6 4 9.36 9 66 23 4 7 9.35 9 68 20 10 2 9.33 9 59 35 3 3 9.25 9 79 19 2 tr 9.22 9 75 21 2 2 9.21 9 78 18 2 tr 2 14.23 14 T20/477112 77 19 1 3 14.22 14 79 19 1 tr 1

Abbreviations: opx orthopyroxene, cpx = clinopyroxene, hb hornblende, bi = biotite, cu curnrningtonite, = titanomagnetite; tr = trace quantities ( %). tm < 1

The ferromagnesian assemblages for tephra samples studied are given in Table 2.2. These can be compared to the dominant mineral assemblages summarised by Froggatt

28 and Lowe (1990). The ferromagnesian assemblage is often useful to narrow down the possible correlatives of a tephra sample, but was rarely able to identify it uniquely. The assemblages are also found to be highly variable due to winnowing effects in distal localities and the common contamination of samples with andesitic tephra minerals. This contamination is evidenced by the presence of abundant clinopyroxene in many of the samples.

2.8.2 Glass chemistry and similarity coefficients

10-22 ka tephras A correlation matrix with the similarity coefficients for the tephra samples from Ruapehu and Tongariro ring plain deposits in the 10-22 ka period with their potential correlatives is presented in Table 2.3. lt has been reported that similarity coefficients for replicate analyses under the same microprobe operating conditions can deviate by around 10 % from the expected ideal value of 1.0 (Froggatt and Solloway, 1986; Stokes et al., 1992). When different beam diameters are used, replicate analyses can have similarity coefficients as low as 0.88 (Pillans and Wright, 1992).

Table 2.3 Correlation matrix of similarity coefficients, comparing glass major oxide chemistry of rhyolitic tephras found in this study with published glass chemistry (Stokes et al., 1992) from potential correlative tephras aged 10-22 ka. Tephra abbreviations from Table 2.1. Sampl e Kp Wh Rr Rk Ok Te

93.5 0.76 0.85 0.81 0.81 0.78 0.80 93.6 0.84 0.91 0.91 0.85 0.82 0.93 93.7 0.78 0.87 0.86 0.82 0.80 0.86 93.8 0.77 0.81 0.82 0.78 0.75 0.82 93.9 0.78 0.92 0.83 0.88 0.87 0.86 93.32 0.80 0.93 0.88 0.88 0.79 0.69 93.59 0.80 0.92 0.88 0.88 0.85 0.88 94.7 0.85 0.94 0.90 0.87 0.84 0.94 94.9 0.94 0.81 0.92 0.75 0.72 0.88 94.21 0.86 0.91 0.93 0.85 0.82 0.94 94.24 0.91 0.77 0.88 0.72 0.70 0.84 94.27 0.80 0.94 0.84 0.94 0.91 0.88 94.28 0.79 0.92 0.84 0.94 0.92 0.86 94.29 0.82 0.90 0.84 0.92 0.91 0.86 94.37 0.86 0.93 0.90 0.87 0.84 0.95 94.47 0.74 0.84 0.77 0.85 0.84 0.81 94.51 0.84 0.94 0.90 0.87 0.84 0.93 94.58 0.81 0.90 0.86 0.87 0.86 0.89 94.71 0.79 0.95 0.84 0.93 0.90 0.87 94.74 0.84 0.93 0.90 0.86 0.83 0.92 94.79 0.85 0.70 0.80 0.67 0.65 0.77 95.2 0.84 0.93 091 0.87 0.84 0.93 95.3 0.79 0.93 0.84 0.93 0.91 0.88 95.6 0.88 0.91 0.93 0.85 0.82 0.96 95.9 0.84 0.90 0.93 0.85 0.81 0.93 95.13 0.86 0.93 0.91 0.87 0.83 0.94

Kp 0.81 0.89 0.76 0.75 0.91 Wh 0.86 0.92 0.89 0.89 Rr 0.80 0.78 0.93 Rk 0.97 0.84 Ok 0.81

29 Most of the tephra samples from this study in the 1 0-22 ka interval have similarity coefficients of 0.90 with published data for two or more of their possible correlatives. 2:: Similarity coefficients comparing published analyses of tephras in the 1 0-22 ka age range (Table 2.3) show a very high degree of correlation between almost all of the possible tephra combinations especially those from the same volcanic centre. The high degree of similarity between the glass chemistry of the tephras of this age range renders the similarity coefficient comparison technique ineffectual and a unique correlation can not be made for each unknown sample.

22-65 ka tephras A correlation matrix of similarity coefficients for tephra samples from Ruapehu and Tongariro ring plain deposits in the 22-65 ka period with their potential correlatives is presented in Table 2.4. Again, most of the samples have more than one potential correlative with the similarity coefficient comparisons. In a comparison of published glass chemistry of the tephras in this time frame (Table 2.4), similarity coefficients indicate a high degree of correlation between all of these tephra combinations with the exception of Hauparu Tephra, which is distinctive. The high degree of similarity between most of the tephra units in this time frame render the similarity coefficient comparisons virtually unusable. This is particularly true for the study area, in which the stratigraphy beneath the Kawakawa tephra was completely unknown prior to this study.

Table 2.4 Correlation matrix of similarity coefficients, comparing glass major oxide chemistry of rhyolitic tephras found in this study with published glass chemistry from potential correlative tephras aged 22-65 ka. Tephra abbreviations from Table 2.1.

Sam[!le Kk* om• Mn* H[!* Re* ot rt 95.1 0.82 0.83 0.82 0.79 0.76 0.78 0.78 93.3 0.90 0.90 0.92 0.89 0.72 0.83 0.85 93.17 0.88 0.88 0.93 0.89 0.73 0.81 0.82 93.33 0.86 0.87 0.88 0.87 0.74 0.82 0.82 7.37 0.85 0.85 0.83 0.82 0.72 0.80 0.81 7.36 0.92 0.90 0.89 0.81 0.70 0.69 0.87 9.37 0.83 0.85 0.83 0.81 0.72 0.78 0.82 9.36 0.92 0.92 0.90 0.81 0.68 0.76 0.87 9.35 0.87 0.87 0.87 0.84 0.75 0.85 0.83 9.33 0.90 0.90 0.89 0.86 0.70 0.87 0.87 9.25 0.73 0.73 0.73 0.74 0.86 0.67 0.69 9.22 0.73 0.75 0.80 0.80 0.85 0.68 0.68 9.21 0.95 0.95 0.93 0.88 0.68 0.87 0.91 14.23 0.77 0.79 0.82 0.83 0.85 0.71 0.72 14.22 0.91 0.93 0.90 0.85 0.68 0.87 0.90

Kk 0.97 0.89 0.85 0.68 0.90 0.90 0.91 0.87 0.69 0.87 0.88 Om0 0.93 0.73 0.84 0.85 Mm 0.77 0.79 0.80 Hp 0.64 0.63 T 0.91

Source of analyses: • Pillans Wright (1992); Froggatt (1982). & t

30 2.8.3 Canonical discriminant function analyses of glass analyses

In both models the transformed oxide values of Si02, Ti02, Al203, FeO, CaO, Na20 and K20 within the glasses were used.

10-22 ka tephras discrimination model The initial discriminant model fo r the six 10-22 ka interval tephras comprises 185 glass analyses. This model can be represented as a plot of the first two (of five) canonical variates calculated from the log-transformed glass analyses (Fig. 2.2A). The mean 2 compositions of each tephra group in the model are presented in Table 2.5. The D statistics indicate good separation between each tephra group with values ranging between 8.9 and 102 (Table 2.6). The classification efficiencies (proportion of correctly identified glass shards based on reclassifying the original data with the calculated discriminant function) of this discriminant model are all high (Table 2.7); the lowest classification efficiency of 89 % is for the Rerewhakaaitu Tephra. Stepwise DFA was attempted for the discriminant model to assess the oxides with the greatest discriminating power. All of the transformed oxides were highly discriminating, in the order of: K20, CaO, Na20, FeO, Si02, Al203 and Ti02• The choice of K20, CaO and FeO near the top of the list of discriminating variables reflect their use by some workers in bivariate or trivariate plots to distinguish TVC and OVC sourced eruptives (e.g., Froggatt and Rogers, 1990; Pillans and Wright, 1992). However, five further oxides also provided discrimination according to STEPDISC, showing that much information is lost when only a subset of the available composition data is used; e.g. compare the plot of CaO vs. FeO for the analyses of the 10-22 ka tephras in Fig. 2.28 to the plot of the first two canonical variates in Fig. 2.2A.

Table 2.5 Mean electron microprobe glass compositions of tephras used in the initial 10- 22 ka discriminant model.

Kara�iti Waiohau Rotorua Rerewhakaaitu Okareka Te Rere

Si02 76.31 (1.08) 78.31 (0.70) 77.72 (0.40) 77.99 (0.33) 78.29 (0.27) 77.36 (0.26) Ti02 0.20 (0.06) 0.14 (0.05) 0.20 (0.04) 0.09 (0.03) 0.12 (0.03) 0.14 (0.04) Al 0 12.98 (0.38) 12.10 (0.27) 12.46 (0.16) 12.45 (0.24) 12.14 (0.13) 12.44 (0.14) 2 3 FeO 1.67 (0.24) 0.91 (0.13) 1.17 (0.15) 0.85 (0.1 1) 0.90 (0.10) 1.19 (0.1 1) M gO 0.17 (0.03) 0.14 (0.03) 0.20 (0.03) 0.08 (0.02) 0.10 (0.03) 0.19 (0.02) CaO 1.39 (0.14) 0.89 (0.12) 1.13 (0.14) 0.68 (0.11) 0.83 (0.08) 1.05 (0.1 1) Na20 3.99 (0.33) 3.99 (0.43) 3.45 (0.26) 3.64 (0.15) 3.60 (0.17) 3.61 (0.13) K20 3.06 (0.24) 3.38 (0.21) 3.50 (0.40) 4.12 (0.29) 3.89 (0.36) 3.85 (0.33) n 45 52 13 27 33 15

31 2.0 .------, B

1.5

"#.

.I:- C> -� 1.0 0 ro + (.) 0.5

I • -6 I 0.0 -6 -4 -2 0 2 4 6 8 0.0L.----1---...1...-----L----1------l 0.5 1.0 1.5 2.0 2.5 Can 1 FeO weight %

Symbol Key 4 c •= Kp •= Wh o = Rr O= Rk + = Ok � = Te

Figure 2.2

A: Plot of the first two canonical variates (Can1 and N Can 2) fo r tephras comprising the initial 1 0-22 ka rhyolitic glass discriminant model; 02 values between groups are given in Table 2.6.

0 8: Plot of FeO vs, CaO weight % within the rhyolitic -2 glass of the tephras comprising the the initial 10-22 ka • discriminant model.

-4 C: Plot of the first two canonical variates (Can 1 and Can 2) fo r tephras comprising the updated 10-22 ka rhyolitic glass discriminant model; 02 values between -6 -4 -2 0 2 4 6 tephra groups are given in Table 2.8. Can 1

4 .------��------�

Symbol Key

O= Kk •= O D= Om + =Hu+=Mg • =T �=Re --- - + N !ii -2 � Figure 2.3 �� () � I (.) �� I Plot of the first two canonical variates (Can 1 and -4 Can 2) fo r tephras comprising the 22-65 ka rhyolitic I glass discriminant model; 02 values between tephra groups are given in table 2.1 1. -6 I + + I

... -6 -3 0 3 6 9 Can 1

32 Table 2.6 02 values and classification efficiencies of the initial discriminant model for the 10-22 ka tephras. Kp Wh Rr Rk Ok T e

values between groups 02Kp 55 17 102 83 43 Wh 24 29 18 20 Rr 65 42 15 Rk 8.9 33 Ok 14 % correctly reclassified by model 98 100 100 89 97 93

Table 2.7 Probabilities of classification of unknown tephras in the 10-22 ka range with their potential correlatives, using the initial 10-22 ka tephra discriminant model. Sample n Overall 12robabili� of classification Kf2 Wh Rr Rk Ok Te

Well classified samples 93.5 8 0.79 0.04 0.17 93.6 11 0.98 0.01 0.01 93.7 9 0.83 0.10 0.07 93.32 9 0.88 0.12 94.7 9 0.78 0.09 0.13 94.9 10 0.87 0.13 94.21 9 0.79 0.03 0.08 0.05 0.05 94.24 10 1.00 94.37 10 0.85 0.14 0.01 94.47 11 0.93 0.07 94.51 9 0.94 0.05 0.01 94.58 8 0.91 0.09 94.74 10 0.99 0.01 94.79 6 1.00 95.2 9 0.98 0.02 95.3 9 0.14 0.82 0.04 95.6 10 0.89 0.09 0.02 95.9 10 0.95 0.04 0.01 95.13 9 0.81 0.19

Poorly classifiedsamples 93.8 14 0.07 0.35 0.26 0.09 0.23 93.9 11 0.61 0.37 0.01 0.01 93.59 10 0.10 0.27 0.09 0.44 0.06 0.04 94.27 9 0.64 0.02 0.23 0.10 0.01 94.28 10 0.01 0.18 0.15 0.30 0.31 0.05 94.29 8 0.01 0.09 0.30 0.29 0.31 94.71 10 0.38 0.1 1 0.33 0.18

n = number of analyses.

Classification of 10-22 ka unknown tephras Twenty-six unknown tephra samples in this time range were tested against the initial discrimination model, 19 were classified with a particular tephra with overall (averaged over the number of analyses for each sample) probabilities exceeding 0.78 (Table 2.7). All of the 19 classifications were consistent with field, stratigraphic and mineralogical evidence. The remaining seven tephra samples had lower overall probabilities of classification to any one potential tephra correlative.

33 An updated discriminant model based on 351 analyses was constructed, with the addition of analyses from the 19 classified tephras to the initial model (Fig. 2.2C). The D2 distances between tephra groups remained high, between 8.7 and 72 (Table 2.8), and so too did the classification efficiencies for each tephra. However, the improvement of this model over the initial one was with its classification of the remaining unknown tephras. This is because the updated model contained more analyses, and thus comprised better samples of the component tephra populations. Also the analyses added to form the updated model were carried out at the same time, by the same operator (SJC) as the remaining unknown samples awaiting classification. This was expected to reduce operator and instrument condition differences which may occur between the unknown analyses and those in the published literature or analysed at an earlier time. Small classification improvements were made with the seven formerly poorly classified tephra samples using the updated model (Table 2.9). Two tephras were well classified,while the remaining five were still poorly classified. The next approach was to examine the possibility of mixed glass populations in the remaining unknown samples by classifying individual shards within each sample. lt appears that each of the five poorly classified tephras contained two glass populations. One population was well classified whilst the other remained poorly classified (Table 2.9). The poorly classified population was later correlated with the Kawakawa Tephra using the 22-65 ka database. The presence of large volumes of Kawakawa Tephra and associated Hinuera Formation (Topping, 197 4) in the study region provides a ready source of Kawakawa glass for admixing with these tephras.

Table 2.8 D2 values and classification efficiencies of the updated discriminant model for the 10-22 ka tephras. Kp Wh Rr Rk Ok Te

values between groups 02Kp 29 14 72 57 29 Wh 18 27 20 17 Rr 55 33 14 Rk 8.7 28 Ok 11 % correctly reclassified by model 96 97 100 84 97 93

22-65 ka tephras discriminant model The initial discriminant model for seven of the largest volume tephras in this age range comprised 110 glass analyses (Fig. 2.3). The mean compositions of each tephra group in the model are presented in Table 2.1 0. The D2 values between tephra groups range between 7.3 and 262 indicating good group separation (Table 2.1 1 ). The closest groups are the Kawakawa and the Okaia Tephras which were erupted from the same volcano within a short space of time (Froggatt and Lowe, 1990). The classification efficiencies of this discriminant model are high for all of the tephra groups; the lowest value of 93% is for the Kawakawa Tephra (Table 2.11) .

34 Stepwise DFA of the analyses of this age group indicated that all of the transformed oxides were highly discriminating. The order in which the oxides were chosen was: K20, FeO, Al203, CaO, Ti02, Si02 and Na20.

Table 2.9 Probabilities of classification of unknown tephras (that were poorly classified by the initial model) in the 10-22 ka range with their potential correlatives, using the updated 10-22 ka tephra discriminant model.

Sample n Overall probabilityof classification Kp Wh Rr Rk Ok Te

Samples with improved classifications 93.9 11 0.95 0.05 94.27 9 0.98 0.01 0.01

Mixed samples* 93.8 14 0.07 0.35 0.26 0.09 0.23 93.8a 8 0.02 0.97 0.01 93.8b 6 0.10 0.15 0.28 0.13 0.34

93.59 10 0.13 0.29 0.04 0.47 0.04 0.03 93.59a 5 0.95 0.05 93.59b 5 0.33 0.47 0.11 0.09

94.28 10 0.01 0.18 0.15 0.30 0.31 0.05 94.28a 6 0.10 0.02 0.06 0.77 0.05 94.28b 4 0.10 0.44 0.05 0.35 0.01 0.05

94.29 8 0.01 0.09 0.30 0.29 0.31 94.29a 4 0.06 0.17 0.77 94.29b 4 0.03 0.28 0.36 0.32 0.01

94.71 10 0.38 0.11 0.33 0.18 94.71a 5 0.01 0.07 0.78 0.14 94.71b 5 0.46 0.12 0.23 0.10 0.09

n = number of analyses. * For mixed samples, the first line is for all shards, (a) denotes the classified subset, and (b) the remaining unclassified subset.

Table 2.10 Mean electron microprobe compositions of tephras used in the 23-65 ka discriminant model.

Kawakawa Okaia Omataroa Mangaone Haupa ru Tihoi Rotoehu

Si02 78.36 (0.47) 78.48 (0.16) 78.17 (0.17) 78.36 (0.13) 76.05 (0.27) 78.47 (0.20) 78.87 (0.49) Ti02 0.14 (0.02) 0.11 (0.03) 0.15 (0.01) 0.17 (0.01) 0.33 (0.04) 0.09 (0.04) 0.13 (0.01) Al203 12.24 (0.16) 12.55 (0.09) 12.30 (0.36) 12.76 (0.05) 13.09 (0.21) 12.53 (0.10) 12.08 (0.19) FeO 1.19 (0.06) 1.15 (0.04) 1.14 (0.05) 1.12 (0.01) 1.75 (0.08) 1.01 (0.09) 0.92 (0.04) MgO 0.13 (0.02) 0.15 (0.01) 0.15 (0.02) 0.19 (0.02) 0.35 (0.05) 0.10 (0.02) 0.12 (0.02) CaO 1.07 (0.07) 1.08 (0.05) 1.07 (0.02) 1.14 (0.04) 1.73 (0.12) 0.79 (0.07) 0.77 (0.06) Na20 3.50 (0.39) 3.45 (0.13) 3.84 (0.09) 3.80 (0.14) 4.09 (0.12) 3.34 (0.26) 3.46 (0.37) K20 3.12 (0.17) 3.00 (0.20) 3.04 (0.21) 2.37 (0.20) 2.50 (0.33) 3.60 (0.13) 3.51 (0.29) n* 68 5 4 4 8 6 15

* each of the averaged analyses is itself a mean of usually 10 analyses.

35 Classification of the 22-65 ka unknown tephras Fifteen unknown tephras were tested against the discriminant model as well as the sub populations of the five mixed tephras mentioned previously from the 1 0-22 ka unknowns. All 15 of the unknown tephras were classified with probabilities of 0.7 or better (Table 2.1 2). All of the tephra samples older than the Kawakawa Tephra were microscopic glass accumulations within weathered andesitic tephras, thus little other corroborating evidence for these correlations was available. However, the correlations produced using the model were in the correct stratigraphic order considering the sampling positions.

Table 2.11 02 values and classification efficiencies of the discriminant model for the 22.5- 65 ka tephras. Kk 0 Om Mg Hu T Re

values between groups 02Kk 7.3 25 77 98 45 64 0 18 59 106 40 61 Om 30 123 50 30 Mg 126 118 85 Hu 262 245 T 35 % correctly reclassifiedby model 93 100 100 100 100 100 100

Table 2.12 Probabilities of classification of unknown tephras in the 22.5-65 ka range with their potential correlatives, using the 22.5-65 ka tephra discriminant model. Sample n Overall probabilityof classification Kk 0 Om Mg Hu T Re

ka unknown tephra samples 93.122.5-65 10 1.00 93.3 10 1.00 93.17 11 0.99 0.01 93.33 11 1.00 7.37 8 0.11 0.89 7.36 6 0.20 0.80 9.37 10 0.71 0.29 9.36 6 0.70 0.30 9.35 10 0.06 0.94 9.33 9 0.14 0.86 9.25 1.00 9.22 1.00 9.21 0.19 0.81 14.23 1.00 14.22 0.10 0.90

Unclassified subsets of mixed ka group samples 93.8b 6 1.00 10-22 93.59b 5 0.98 0.02 94.28b 4 0.73 0.27 94.29b 4 1.00 94.71b 5 0.70 0.30

n = number of analyses.

The unknown analyses not classified by the 10-22 ka model in the five mixed tephras were classified as Kawakawa Tephra with probabilities >0.7. This indicates that the five

36 mixed tephras were contaminated by reworked Kawakawa Tephra shortly after their deposition. The tephras were being deposited between 21.1 and 14.7 ka B.P., during a period when the climate was cool, dry and stormy (McGione and Topping, 1983). Much erosion and deposition of ring plain deposits (including the Oruanui lgnimbrite member of the Kawakawa Tephra) by lahars and fluvial activity was occurring at this time on the ring plain (Cronin and Neall, in press). These conditions probably led to much Kawakawa Tephra glass being transported aerially and fluvially to contaminate the tephras and other ring plain deposits of this age.

2.9 Conclusions

In this study we have extended the scope of past DFA work and show a practical example of these statistical methods in action. From this work we have reached the following conclusions: 1) The creation of canonical discriminant models for potential tephra correlatives in two clearly defined stratigraphic intervals offers a considerable improvement on other methods for the identification of the tephras in this study. 2) Most unknown tephras in the two stratigraphic groups were classified using the initial discriminant models based on past analyses of the potential correlative tephras. However, small improvements in classification of unknown tephras were achieved by including correctly classified tephras into the model. This should only be done where all other corroborating evidence supports the statistical classification. The improvements were probably due to larger sample sizes in the updated model, and a reduction of differences in operators and instrument conditions between the unknown tephras being tested and the past analyses making up the initial model. 3) DFA enables the use of all of the glass analyses of a particular unknown tephra to be used to classify or identify it. We have shown this to be of great value for identifying mixed tephras in samples containing mixed glass populations. This approach will be especially practical when examining tephras which occur as microscopic concentrations in slowly accumulating sediments such as loess and aggrading soils. In these sediments there is a common mixing upwards of glass grains so that younger tephras need to be separated from a background of older shards into which they are mixed (Eden et al., 1992). 4) In this study we have identified four previously undiscovered rhyolitic tephras in this area from glass concentrations within buried soil and andesitic ash. These are the Okaia, Omataroa and Hauparu Tephras and the Rotoehu Ash; they assist significantly in establishing a chronology for andesitic tephras and diamictons older than 22.6 ka B.P.

2.1 0 Acknowledgements

SJC gratefully acknowledges funding from the New Zealand Vice-Chancellor's Committee, Massey University Graduate Research Fund, the Helen E. Akers Scholarship

37 Fund, and the Department of Soil Science of Massey University. Ken Palmer (Victoria University of Wellington) is thanked for his introduction to the operation of the electron microprobe. Comments from two journal reviewers helped us to clarify our arguments, for which we are grateful.

2.11 References

Aitchison, J., 1983. Principal component analysis of compositional data. Biometrika 70: 57-65. Aitchison, J., 1986. The statistical analysis of compositional data. Monographs on Statistics and Applied Probability. Chapman and Hall, London. Alloway, B.V., 1989. Late Quaternary cover-bed stratigraphy and tephrochronology of north-eastern and central Taranaki, New Zealand. Unpublished Ph.D. thesis, Massey University, Palmerston North. Alloway, B.V.; Neall, V.E.; Vucetich, C.G., 1992. Particle size analyses of Late Quaternary allophane-dominated andesitic deposits from New Zealand. Quat. Inter. 13/14: 167-1 74. Beaudoin, A.B.; King, R.H., 1986. Using discriminant function analysis to identify Holocene tephras based on magnetite composition: a case study from the Sunwapta Pass area, Jasper National Park. Canadian J. of Earth Sci. 23: 804- 812. Blakemore, L.C.; Searle, P.L.; Daly, B.K., 1987. Methods for chemical analysis of soils. N. Z. Soil Bureau Scientific Rep. 80: 103pp. Borchardt, G.A .. ; Harward, M.E.; Schmitt, R.A., 1971 . Correlation of ash deposits by activation analysis of glass separates. Quat. Res. 1: 24 7-260. Carter, L.; Nelson, C.S.; Neil, H.L.; Froggatt, P.C., 1995. Correlation, dispersal, and preservation of the Kawakawa Tephra and other late Quaternary tephra layers in the Southwest Pacific Ocean. N. Z. J. of Geol. and Geophys. 38: 29-46. Cronin, S.J.; Neall, V.E., in press. A Late Quaternary stratigraphic framework for the northeastern Ruapehu and eastern Tongariro ring plains, New Zealand. N. Z. J. of Geol. and Geophys. 40. Cronin, S.J.; Neall, V.E.: Palmer, A.S., 1996b. Geological history of the northeastern ring plain of Ruapehu volcano, New Zealand. Quat. Inter. 34-36: 21-28. Cronin, S.J.; Neall, V.E., Palmer, A.S., 1996c. Investigation of an aggrading paleosol developed into andesitic ring-plain deposits, Ruapehu volcano, New Zealand. Geoderma 69: 119-135. Cronin, S.J.; Neall, V.E.; Stewart, R.B.; Palmer, A.S., 1996a. A multiple-parameter approach to andesitic tephra correlation, Ruapehu volcano, New Zealand. J. Volcano!. and Geotherm. Res., 72: 199-215. Donoghue, S.L., 1991. Late Quaternary volcanic stratigraphy of the southeastern sector of Mount Ruapehu ring plain, New Zealand. Unpub. Ph.D. thesis, Massey University, Palmerston North.

38 Donoghue, S.L.; Neall, V.E.; Palmer, A.S., 1995. Stratigraphy and chronology of late Quaternary andesitic tephra deposits, Tongariro Volcanic Centre, New Zealand. J. Roy. Soc. N. Z. 25: 115-206. Eden, D.N.; Froggatt, P.C.; Mclntosh, P.D., 1992. The distribution and composition of volcanic glass in late Quaternary loess deposits of southern South Island, New Zealand, and some possible correlations. N. Z. J. of Geol. and Geophys. 35: 69- 79. Ewart, A., 1963. Petrology and petrogenesis of the Quaternary pumice ash in the Taupo area, New Zealand. J. Petrology 4: 392-431. Froggatt, P.C., 1982. A study of some aspects of the volcanic history of the Lake Taupo area, North Island, New Zealand. Unpub. Ph.D. thesis, Victoria University of Wellington, Wellington. Froggatt, P.C., 1983. Towards a comprehensive Upper Quaternary tephra and ignimbrite stratigraphy in New Zealand using electron microprobe analysis of glass shards. Quat. Res. 19: 188-200. Froggatt, P.C.; Solloway, G.J., 1986. Correlation of Papanetu Tephra to Karapiti Tephra, central North Island, New Zealand. N. Z. J. of Geol. and Geophys.29: 303-313. Froggatt, P.C.; Lowe, D.J., 1990. A review of late Quaternary silicic and some other tephra formations from New Zealand: their stratigraphy, nomenclature, distribution, volume, and age. N. Z. J. of Geol. and Geophys. 33: 89-109. Froggatt, P.C.; Rogers, G.M., 1990. Tephrostratigraphy of high altitude peat bogs along the axial ranges, North Island, New Zealand. N. Z. J. of Geol. and Geophys. 33: 111-124. Hackett, W.R.; Houghton, B.F., 1989. A facies model for a Quaternary andesitic composite volcano: Ruapehu, New Zealand. Bull. Volcano!. 51: 51-68. Howorth, R., 1975. New formations of Late Pleistocene tephras from Okataina Volcanic Centre, New Zealand. N. Z. J. of Geol. and Geophys. 18: 683-712. Johnson, R.A.; Wichern, D.W., 1992. Applied multivariate statistical analysis (3rd Edition). Prentice-Hall lnc., New Jersey. Kohn, B.P., 1970. Identification of New Zealand tephra layers by emission spectrographic analysis of their titanomagnetites. Lithos 3: 361 -368. Lowe, D.J., 1988. Stratigraphy, age, composition, and correlation of late Quaternary tephras interbedded with organic sediments in Waikato lakes, North Island, New Zealand. N. Z. J. of Geol. and Geophys. 31 : 125-165. Mathews, W.H., 1967. A contribution to the geology of the Mount Tongariro massif, North Island, New Zealand. N. Z. J. of Geol. and Geophys. 10: 1027-1038. McGione, M.S.; Topping, W.W., 1983. Late Quaternary vegetation, Tongariro region, central North Island, New Zealand. N. Z. J. of Botany 21: 53-76. Pillans, B.J.; Wright, I., 1992. Late Quaternary tephrostratigraphy from the southern Havre Trough - Bay of Plenty, northeast New Zealand. N. Z. J. of Geol. and Geophys. 35: 129-143.

39 Pillans, B., McGione, M., Palmer, A., Mildenhall, D., Alloway, B.; Berger, G., 1993. The Last Glacial Maximum in central and southern North Island: a paleoenvironmental reconstruction using the Kawakawa tephra formation as a chronostratigraphic marker. Palaeo., Palaeo., Palaeo. 101: 283-304. Randle, K.; Gorton, G.G.; Kittleman, L.R., 1971. Geochemical and petrological characterisation of ash samples from Cascade Range volcanoes. Quat. Res. 1: 261 -282. SAS Institute Inc., 1989. SAS users guide: statistics. Version 6 Edition. Cary N.C., SAS Institute Inc., 1028 pp. Shane, P.A.R., Froggatt, P.C., 1994. Discriminant function analysis of glass chemistry of New Zealand and North American tephra deposits. Quat. Res. 41: 70-81. Smith, D. G. W.; Westgate, J.A., 1969. Electron probe technique for characterising pyroclastic deposits. Earth Planet. Sci. Lett. 5: 313-319. Srivastiva, M.S.; Carter, E.M., 1983. An introduction to applied multivariate statistics. Elsevier Science Publishing Co. Inc., New York. Stokes, S.: Lowe, D.J., 1988. Discriminant function analysis of late Quaternary tephras from five volcanoes in New Zealand using glass shard major element chemistry. Quat. Res. 30: 270-283. Stokes, S.; Lowe, D.J.; Froggatt, P.C., 1992. Discriminant function analysis and correlation of Late Quaternary rhyolitic tephra deposits from Taupe and Okataina volcanoes, New Zealand, using glass shard major element composition. Quat. lnter. 13/14: 103-1 17. Topping, W.W., 1974. Some aspects of the Quaternary history of the Tongariro Volcanic Centre, New Zealand. Unpub. Ph.D. thesis, Victoria University of Wellington, Wellington. Topping, W.W.; Kohn, B.P., 1973. Rhyolitic tephra marker beds in the Tongariro area, North Island, New Zealand. N. Z. J. of Geol. and Geophys. 16: 375-395. Vucetich, C.G.; Pullar, W.A., 1969. Stratigraphy and chronology of late Pleistocene volcanic ash beds in central North Island, New Zealand. N. Z. J. of Geol. and Geophys. 12: 784-837. Vucetich, C.G.; Pullar, W.A., 1973. Holocene tephra formations erupted in the Taupe area and interbedded tephras from other volcanic sources. N. Z. J. of Geol. and Geophys. 16: 745-780. Vucetich, C.G.; Howorth, R., 1976. Late Pleistocene tephrostratigraphy in the Taupe district, New Zealand. N. Z. J. of Geol. and Geophys. 19: 51-69. Wilson, C.J.N.; Switzur, R.V.; Ward, A.P., 1988. A new 14C age for the Oruanui (Wairakei) eruption, New Zealand. Geol. Mag. 125: 297-300. Wilson, C.J.N.; Houghton, B.F.; Lanphere, M.A.; Weaver, S.D., 1992. A new radiometric age estimate for the Rotoehu Ash from Mayor Island volcano, New Zealand. N. Z. J. of Geol. and Geophys. 35: 371-374.

40 CHAPTER 3. ANDESTIC TEPHROCHRONOLOGY

3.1 Introduction

The work of Topping (1973, 1974) and Donoghue et al. (1995) has resulted in a well defined stratigraphy of Ruapehu- and Tongariro-sourced tephras since 22.6 ka B.P (Chapter 1: Table 1.3). The identification of these tephras in the ring plain area relies mostly on their physical appearance in conjunction with their stratigraphic position. In more distal areas the identification of TgVC-sourced tephras relies on other features such as characteristic mineralogy as well as chemistry of glass or phenocryst minerals (e.g. Lowe, 1988a; Donoghue et al., 1991 ). There were two ways in which this study was intended to add to previous andesitic tephrochronology work in the study area as well as to andesitic tephrochronolgy as a whole. The first contribution was to attempt additional methods of andesitic tephra identification such as the use of the major oxide chemistry of ferromagnesian phenocryst minerals within the tephras. To compare the chemical compositions of the tephras in this study, statistical methods similar to those used for the rhyolitic tephras in Chapter 2 were used. The second contribution of this study was to investigate the pre-22.6 ka B.P. andesitic tephrochronology of the NE Ruapehu and E Tongariro ring plains. These two contributions have been written up as the following two papers. In the first study, the major oxide chemistry of hornblende and titanomagnetite phenocrysts of TgVC- and Egmont volcano (EV)-sourced tephras were used to test and develop the canonical discriminant function analysis (DFA) method. Once it was determined whether this method was able to distinguish between tephras from the two sources, discrimination between individual tephras from the same source was attempted. A well controlled sequence of EV-sourced tephras was chosen for this trial. The second study involved combining many parameters to correlate the pre-22.6 ka B.P. sequence of andesitic tephras on the NE Ruapehu ring plain. Physical appearance, mineralogy and mineral chemistry were all used in combination to correlate this sequence. A statistical clustering method was used to establish tephra groupings which could then be discriminated using the DFA methods developed in the first study. A listing of all electron microprobe analyses used in these studies comprises Appendix 3. The SAS programs used in the two studies are included in Appendix 1.

3. 1. 1 Photographs of the Upper Wa ikato Stream sequence

The following plates display part of the Upper Waikato Stream sequence and three of the marker units described on the basis of their appearance in Section 3.4. These plates were not included in either of the papers comprising this chapter.

41 Plate 3.1 . Part of the Upper Waikato Stream sequence located at T20/468102 and represented in Figs. 3.6, 4.1 and 5.3. The exposure is ea. 30 m high.

position of Kawakawa Te phra

R11 lahar deposits (Fig. 5.3)

position of Marker Unit 2 (Fig. 3.6)

R12 lahar deposits (Fig. 5.3)

Marker Unit 3 (Fig. 3.6) indicated by arrow on the left of photograph.

Plate 3.2. Marker Unit 1 tephra at T20/464098, see Fig. 3.6 for stratigraphic position and Section 3.4.6 for description. Lens cap is 50 mm in diameter. 42 Plate 3.3. Marker Unit 2 tephra at T20/4691 00; see Fig. 3.6 for stratigraphic position and Section 3.4.6 for description. Lens cap is 50 mm in diameter.

Plate 3.4. Marker Unit 3 tephra at T20/4681 02; see Fig. 3.6 for stratigraphic position and Section 3.4.6 for description. Lens cap is 50 mm in diameter.

43 3.2 Contributions of co-authors

Sourcing and identifying andesitic tephras using major oxide titanomagnetite and hornblende chemistry, Egmont volcano and Tongariro Volcanic Centre, New Zealand

Bulletin of Volcanology, 58: 33-40 (1996).

S.J. Cronin: Principal investigator Carried out: sampling and laboratory preparation of TgVC tephras electron microprobe analysis of TgVC tephras all statistical programming and analysis manuscript preparation and writing

R.C. Wallace: Go-investigator Carried out: sampling and laboratory preparation of EV tephras electron microprobe analysis of EV tephras editing and discussion of the manuscript

V.E. Neall: Adviser: aided the study by discussing results and methodology as well as editing and discussion of the manuscript.

A multiple-parameter approach to andesitic tephra correlation, Ruapehu volcano, New Zealand

Journal of Volcanology and Geothermal Research, 72: 199-215 (1996).

S.J. Cronin: Principal investigator Carried out: sampling and laboratory preparation of tephras optical microscopy electron microprobe analysis statistical programming and analysis manuscript preparation and writing

V.E. Neall, R.B. Stewart, A.S. Palmer: Advisers aided the study by discussing results and methodology with SJC as well as editing and discussion of the manuscript.

44 3.3 SOURCING AND IDENTIFYING ANDESITIC TEPHRAS USING MAJOR OXIDE TITANOMAGNETITE AND HORNBLENDE CHEMISTRY, EGMONT VOLCANO AND TONGARIRO VOLCANIC CENTRE, NEW ZEALAND

S. J Cronin, R C. Wallace, and V. E Neall Department of Soil Science, Massey University, Private Bag 11 222, Palmerston North, New Zealand

1996, Bulletin of Volcanology, 58: 33-40.

Abstract. Canonical discriminant function analysis was employed to discriminate between electron microprobe-determined titanomagnetite and hornblende analyses from Egmont volcano and Tongariro Volcanic Centre. Data sets of 436 titanomagnetite and 206 hornblende analyses from the two sources were used for the study. Titanomagnetite chemistry provided the best discrimination between these two sources with classification efficiencies of 99 % for sample averages and 95 % for individual analyses. The difference between sources for hornblende chemistry was less marked, but classification efficiencies of 1 00 % for sample averages and 87 % for individual analyses were achieved. Using the same methods, a preliminary discrimination of individual Egmont volcano-sourced tephras was attempted. Titanomagnetite chemistry enabled the discrimination of several individual tephras or at least pairs of tephra units, but hornblende chemistry provided little discrimination. This technique provides an improvement on previous methods for chemically distinguishing distal tephra from the two sources as well as potentially identifying individual tephras from a particular source. A major advantage over previous discrimination techniques is that individual analyses can be classified with a known probability of group membership (with groups such as volcano source or an individual tephra unit). Tephras in a depositional environment where mixing is common such as within soil, loess and marine sequences, can be sourced or identified more easily with classification of individual grains.

Key words: Tephrostratigraphy-Egmont volcano-Tongariro Volcanic Centre- discriminant function analysis-OFA.

3.3.1 Introduction

The stratigraphic importance of tephra layers is well recognised in Quaternary studies world-wide as well as in New Zealand (e.g. Thorarinsson, 1949; Pullar, 1967; Lowe, 1990). Tephrochronology studies in New Zealand have mostly concentrated on the many, widespread and voluminous rhyolitic tephras erupted from calderas within the Taupo Volcanic Zone (e.g. Vucetich and Pullar, 1969 and 1973; Froggatt, 1983; Froggatt and Lowe, 1990). Studies of andesitic tephrochronology have been fewer in New Zealand due to the more restricted distribution of andesitic tephras, the ease with which

45 they weather, apparent difficulties involved in their identification, and their perceived limited use in petrographic studies. Andesitic tephrochronology has however proved to be important in studies on and around the two major andesitic volcanic areas in the North Island (Fig. 3.1 ); Egmont volcano and Tongariro Volcanic Centre (e.g. Kohn and Neall, 1973; Topping, 1973; Donoghue et al., 1995), as well as in distal regions (Lowe, 1988a; Wallace, 1987; Eden et al. , 1993). In these areas andesitic tephras have been used to either complement the rhyolitic tephra record or provide greater stratigraphic resolution.

175" E

38° s

0 50 100 km

Fig 3.1. North Island of New Zealand with locations of Egmont volcano and Tongariro Volcanic Centre.

Methods of rhyolitic tephra identification have become increasingly sophisticated as the need arises for accurate and reliable identification of tephras in distal areas, where field criteria are sometimes ambiguous. Methods have involved mineralogical criteria, (e.g. Randle et al. , 1971; Lowe, 1988) as well as chemical criteria, (e.g. Smith and Westgate, 1969; Kohn, 1970; Borchardt et al., 1971; Froggatt, 1983; Wallace, 1987; Stokes et al., 1992). Chemical techniques have mostly involved glass and titanomagnetite, and in New Zealand glass chemistry has achieved the most widespread use for chemical characterisation of tephra (e.g. Froggatt, 1983; Stokes et al. 1992). Andesitic tephra identification methods in New Zealand have followed those developed for rhyolitic tephras. However in distal areas of tephra accumulation it remains difficult to distinguish Tongariro Volcanic Centre- (TgVC) and Egmont volcano- (EV) sourced tephras because they have similar mineralogy, and mixing occurs. The greatest difficulty with the identification of andesitic tephras results from them being more easily

46 weathered in most depositional environments than rhyolitic tephras (Kirkman and McHardy, 1980; Lowe 1986). There are major differences in glass chemistry between the two centres, (Wallace et al., 1986) but rapid weathering, renders andesitic glass unusable for tephra identification in most circumstances. Mineralogy is often but not always useful for distinguishing EV- and TgVC-sourced distal tephras. Lowe (1988a) distinguished EV tephras by a ferromagnesian assemblage of clinopyroxene + hornblende ± orthopyroxene, compared with orthopyroxene clinopyroxene ± olivine ± hornblende for + TgVC tephras. However, Wallace et al. (1986) concluded that olivine and orthopyroxene are more abundant in EV-sourced tephras than previously thought, and several TgVC sourced tephra layers also lie outside of the general assemblage fields defined by Lowe (Donoghue et al. , 1991; Cronin et al. , 1996). Mineral assemblages may also be affected

by winnowing (Juvigne and Shipley, 1983), weathering (Hodder et al., 1991 ) , and pedogenic mixing (Wallace, 1987). Titanomagnetite trace element chemistry has been shown to be useful in distinguishing individual EV tephras, (Kohn and Neall, 1973; Wallace et al. , 1986) as well as distinguishing between EV- and TgVC-sourced tephras (Kohn and Neall, 1973; Lowe, 1988a). Kohn and Neall (1973) used emission spectrographic analyses on bulk samples of titanomagnetites, but this technique is unsuitable in any environment where there is potential mixing of tephra layers, such as in loess and soil. The elements which provided the greatest discrimination of sources were, chromium, vanadium, nickel, and manganese. Lowe (1988a) used a bivariate plot of Cr203 vs. MnO wt % to show a provisional difference between tephras from the two sources. Hornblende chemistry has also been used to distinguish between EV- and TgVC sourced tephras. Wallace (1987) used MgO, FeO and K20 content of hornblende to distinguish between EV- and Taupe Volcanic Zone-sourced tephras and Eden et al. (1993), used calcium and silicon to discriminate between TgVC and EV hornblendes. These approaches may be useful but they do not use the full range of chemistry data available. Other mineral phases which have been used to discriminate individual tephras or provide limited discrimination between EV and TgVC sources are olivine, clinopyroxene, and plagioclase (Lowe, 1988a; Donoghue et al. , 1991 ). Compared to bivariate plots, a more statistically powerful technique which uses the total chemistry of a sample in comparing and contrasting compositional data is canonical discriminant function analysis (DFA). DFA was introduced to tephra studies by Borchardt et al. (1971) using rhyolitic glass analyses, and has since been applied to titanomagnetite major element analyses to discriminate tephras in Canada (King et al., 1982; Beaudion and King, 1986). In New Zealand, DFA using major oxide glass analyses has been used to discriminate volcanic source areas as well as some individual rhyolitic eruptives (Stokes and Lowe, 1988; Stokes et al., 1992; Shane and Froggatt, 1994). This study was firstly aimed at discriminating between TgVC- and EV-sourced andesitic tephras by using canonical DFA of all titanomagnetite and hornblende major oxides. The second step was to apply this methodology on the same data set in a preliminary investigation to discriminate individual tephra eruptives from one of the

47 sources. We also present here preliminary results from our investigation of EV-sourced tephras.

3.3.2 Methods

Tephra samples were taken from localities proximal to EV and TgVC. Phenocryst phases were extracted, mounted in epoxy resin, cut and polished. Phenocrysts of titanomagnetite and hornblende were analysed in this study. Orthopyroxene, clinopyroxene and plagioclase were not used because they display strong oscillatory zoning in TgVC tephras and the chemical variation within a single grain was as great as the variation between samples. Olivine was also not targeted because it was not present in many of the tephras. The chemistry of the titanomagnetites and hornblendes varied less within each sample population and there was no observable oscillatory zoning. Further advantages of using titanomagnetite are that it is relatively stable during weathering (Ruxton, 1968) and is easy to separate from a sample. The analyses were carried out with a JEOL JXA-733 electron microprobe using an accelerating voltage of 15 kV, a 12 nA beam current, and a 3 11m beam diameter. Mineral standards of the Victoria University of Wellington collection were analysed routinely to check for any machine drift. Between six and ten titanomagnetite and hornblende crystals were analysed where possible from each tephra sample.

3.3.3 Statistical Methods

Canonical discriminant function analyses (DFA) were used to discriminate between the Egmont and Tongariro mineral analyses. This is a technique related to principle component analysis that reduces the dimensionality of data, such as compositional analyses, which consist of a large number of variables. Canonical DFA enables the derivation of a small number of linear combinations of the quantitative variables which best discriminate pre-defined groups of observations or analyses. This means that instead of working with ten oxide scores to discriminate groups of samples, one or two canonical variables contain the information. Before identification of an unknown tephra can be undertaken, a reference set of data with known associations must be established. The canonical DFA from this known set is used to produce a discriminant model which then can be used to classify unknown data. The analyses obtained in this study from EV and TgVC enable the creation of a discriminant model. The theory of OFA has been outlined by many authors (e.g. Srivastava and Carter, 1983; Johnson and Wichern, 1992) but is not discussed further here. As part of canonical DFA the Mahalanobis distance statistic or 02 value is often calculated. lt indicates the multivariate spacing between data groups in multiple dimensions (Srivastava and Carter, 1983). The 02 statistic is therefore a measure of the separation of groups of samples (Stokes et al. , 1992), in a similar manner to the coefficients of variation and similarity coefficients of Borchardt et al. (1971 ).

48 Beaudion and King (1986), and Stokes and Lowe (1988) used a technique which selected variables that had the greatest discriminating power. This subset of highly discriminating variables was shown to improve classification within groups and discrimination between groups. This method is known as stepwise DFA and is further described by Srivastava and Carter (1983), and Stokes and Lowe (1988). In this study the SAS system programs STEPDISC and CANDISC were used (SAS Institute Inc., 1985). These programs perfo rm stepwise DFA and canonical DFA respectively. Prior to statistical analysis the data used in this study were transformed in the manner described by Aitchison (1983) and Stokes and Lowe (1988). This is necessary due to the statistical problem of closure in compositional data. The pre-treatment used is termed a log ratio transformation, whereby one oxide score is used to divide into all of the other scores and the logarithm (base 1 0) is taken of each of these ratios. Use of the logarithm constrains component scores to being greater than zero.

3.3.4 Discrimination between sources

Titanomagnetite In this study 300 titanomagnetite phenocrysts were analysed from 70 individual TgVC sourced tephra units, and 136 from 11 individual EV-sourced tephra units. The tephra units sampled range in age from ea. 3-35 ka. The mean titanomagnetite compositions for each centre are presented in Table 3.1. The analyses were normalised and then transformed with the log-ratio transformation using Si02 as the oxide divisor. The choice of oxide for the divisor was found to have no effect on the discriminating ability of the DFA (Stokes and Lowe, 1988).

Table 3.1. Titanomagnetite and hornblende average chemistry from the two tephra sources. Titanomagnetite Hornblende Oxide wt% Tongariro Egmont Tongariro Egmont

Si02 0.18 (0.16) 0.18 (0.06) 43.13 (1 .55) 42.62 (1 .33) Ti02 10.66 (2.22) 9.09 (1 .40) 2.22 (0.65) 3.35 (1.95) Al203 4.00 (1 .10) 3.91 (1 .32) 12.46 (1 .43) 11.99 (2.30) FeO 80.81 (2.02) 82.65 (1 .79) 12.77 (1.17) 11.99 (0.83) M nO 0.42 (0.19) 0.68 (0.20) 0.33 (0.15) 0.25 (0.13) M gO 3.13 (0.68) 3.35 (0.65} 14.51 (0. 78) 14.27 (0.66) CaO nd nd 11.58 (0.62) 12.06 (0.39) Na20 nd nd 2.34 (0.34) 2.50 (0.12) Kp nd nd 0.58 (0.27) 0.96 (0.09) Cr203 0.42 (0.38) 0.07 (0.07) nd nd n 300 136 117 89 n = number of analyses to calculate the mean, standard deviations in brackets. nd = not determined

49 Two SAS data sets were created for OFA, one using the mean analyses for each tephra unit, and one using all of the individual analyses. Shane and Froggatt (1994) found that mean analyses produced better discrimination than individual glass shard analyses for New Zealand rhyolitic tephras. The total spread of the data is however better indicated by individual analyses. The variables used in both databases were transformed Ti02, Al203, FeO, MnO, MgO and Cr203. Canonical OFA of the means data set produced very good separation between the EV and TgVC sample groups. A plot of the first two canonical variates is 2 used to illustrate the separation (Fig. 3.2A). The 0 value of 18.1 between groups indicates a good separation of the two groups with very little overlap occurring. The efficiency of group classification is presented in Table 3.2. The stepwise procedure STEPOISC selected the most discriminating variables in order of Cr203, MnO, Ti02, FeO and Al203. Using these variables it was possible to achieve a group separation almost as good as using all of the oxides. The Cr203 values in the EV-sourced tephras are very low, close to detection limits of the microprobe. Removing Cr203 from the OFA reduces the 2 degree of separation of the two tephra groups to a 0 value of 12.3 (Fig. 3.28, Table 3.2). The most discriminating variables now become MnO, FeO, Ti02, and MgO. Canonical OFA using all of the data including Cr203 also produced very good group separation (Fig. 3.2C). There is very little overlap between the two groups and the 2 0 value of 13.1 is still high indicating good separation (Table 3.2). The STEPOISC procedure selected the most discriminating variables in order of Cr203, MnO, Ti02, MgO, 2 and FeO. Removing Cr203 from the OFA reduced the 0 value between the source groups to 7.2 but still enabled a good classification of the groups (Table 3.2). The most discriminating variables became MnO, FeO, Ti02, and MgO.

Hornblende 117 hornblende analyses were obtained from 23 individual TgVC-sourced tephras, and 89 analyses from 11 individual EV-sourced tephras. A further 15 hornblende analyses from 5 EV-sourced tephras were also available (Lowe, 1987). The mean hornblende compositions for each centre are presented in Table 3.1. The analyses were normalised and transformed with the log-ratio transformation using Na20 as the oxide divisor. As for titanomagnetite, two SAS data sets were created, one with the average analysis for each tephra, the other with all of the analyses. The same transformed oxides were used in both data sets: Si02, Ti02, Al203, FeO, MnO, MgO, CaO, and K20. Canonical OFA of the average data produced a good separation of the TgVC and 2 EV tephras (Fig. 3.3A). The 0 value of 9.9 is not as high as for the titanomagnetite averages separation, but still indicates the volcanic centre groups are well separated with no overlap (Table 3.2). The STEPOISC procedure selected the most discriminating variables in order of K20, MnO, Si02, CaO, and MgO.

50 6 4 A. A.

+ 0 + 3 2 0 N N 0 8 0 - � 0 �0 J) {) {) o c9Cfj 0

-3 -2

-6 +----r---1------.----.-----, -4 +----.-----1,---.,----..., -4 -2 0 2 4 6 -4 -2 0 2 4 Can 1 Can 1 6 6 + B. B. + 4 + 3

2 N N c:: � 0 Cll {) {) 0

-3 -2

-4 +----r----+---.-----, -4 -2 0 2 4 6 -4 -2 0 2 4 Can1 Can 1 6 c. Figure 3.3 Plots of the first and second canonical variates (Can 1 2 + -lt and Can2) for hornblende data together with D values 3 between source groups. Circles represent EV and Crosses TgVC data. (A) Sample mean data, (B) all + individual analyses. N � 0 - � {) 0

-3 +

-6 +----..--�--+--....----r-----, -6 -4 -2 0 2 4 6 Can 1 Figure 3.2 Plots of the first and second canonical variates (Can 1 and Can 2) for titanomagnetite data together with values between source groups. Circles represent EV02 data and Crosses TgVC data. (A) Sample mean data with all oxide variables, (B) sample mean data excluding Cr,03, (C) all individual analyses using all oxides.

51 Canonical OFA using all of the individual analyses produced a reasonable separation of samples from the two sources (Fig. 3.38). There is a small degree of overlap between 2 groups and the 0 value is correspondingly lower at 4.4 (Table 3.2). The STEPOISC procedure selected all of the variables as being highly discriminating but in the order of K20, MnO, Si02, MgO, FeO, Ti02, Al203, and CaO.

Table 3.2. Classification efficiency of the between-source discriminant functions.

Actual group Number of Classified group membership Overall observations classification TgVC EV efficiency (%)

Titanomagnetite Mean data TgVC 70 69 1 EV 11 11 98.8 Mean data excluding Cr2 03 TgVC 70 65 5 EV 11 11 93.8 Individual data TgVC 300 285 15 EV 136 7 129 95.0 Individual data excluding Cr203 TgVC 300 277 23 EV 136 12 124 91.9

Hornblende Mean data TgVC 23 23 EV 16 16 100.0 Individual data TgVC 117 93 24 EV 89 2 87 87.4

3.3.5 Discrimination of individual Egmont volcano-sourced tephras

Titanomagnetite An EV-source subset of the individual titanomagnetite analyses database described in the previous section was used to attempt discrimination of individual tephras. The analyses from each tephra unit in Table 3.3 were grouped separately except for samples Tariki a, Tariki b,c,d and, Tariki e, f which were combined as a single group representing Tariki tephra. The separation of these tephra units appeared very promising with the canonical 2 OFA approach (Fig. 3.4A, Table 3.4). 0 distances between many of the tephras were very high and their degree of classification was also good (Table 3.4). Four tephra units were not as well separated from all of the others; Konini Tephra and Kaponga Tephra analyses appear very closely related, also Mahoe Tephra and Poto Tephra (members f, g) were also very closely related. The most discriminating variables were chosen as MnO, MgO, FeO, Ti02 and Al203.

52 Table 3.3. Egmont-sourced tephras used in the discrimination study. Tephra unit Symbol Estimated age (AIIoway et al., 1995) (ea. ka B.P.)

Manganui Tephra Mg 2.9 -3.3 lnglewood Tephra 11 3.6 -3.7 Mangatoki Tephra Mt 4.1 -4.6 Tariki Tephra, e, f l Tariki Tephra, d, c, b � Tr 4.6 -5.2 Tariki Tephra, a J Waipuku Tephra Wp 5.0 - 5.2 Kaponga Tephra Kp 5.3 - 10.0 Konini Tephra Kn 10.1 -10.4 Mahoe Tephra Mh 10.4 - 12.0 Poto Tephra, f, g Pt 22.5

Hornblende An EV-source subset of the database of hornblende individual analyses was used for this discrimination. The analyses from each tephra unit in Table 3.3 were grouped in the same manner as for the titanomagnetite data. The DFA enabled very little discrimination of the individual tephra units (Fig 3.48). The 02 distances were very low between many of the tephra sample groups, and within many tephra groups there was a large spread of data. The classification efficiency was also very low with this data (Table 3.4). The only tephra unit which was effectively discriminated was the Manganui tephra.

Table 3.4. Classification efficiencies of the discriminant function analyses for the individual Egmont-sourced tephras. Actual Number of Classified group membership Overall group* observations classification Mg 11 Mt Tr Wp Kp Kn Mh Pt efficiency (%)

Titanomagnetite Mg 11 11 11 4 2 2 Mt 10 10 Tr 38 4 28 6 83.3 Wp 11 1 10 Kp 6 6 Kn 13 4 9 Mh 12 11 1 Pt 9 1 8

Hornblende Mg 3 3 11 9 6 2 Mt 7 2 3 1 Tr 20 2 2 7 1 2 4 Wp 8 1 1 6 55.9 Kp 4 3 Kn 4 3 Mh 5 1 3 Pt 8 1 5

* Tephra abbreviations from Table 3.3.

53 6 3 A. B . 0 4 2 1 6 o l .\. X+ �0 +� I • X 2 1 � 6 • • 0 ' N • L. N ------x r c: roc: 0 ro 0 tJ3 • :+�\;!6 \00 () : () X Q -2 0 -1 Di 6tf� � �0 T • o1 T oo I -4 : -2 . T o s 0 I 4 • • -6 I -3 I -10 -5 0 5 10 15 -6 -4 -2 0 2 4 Can 1 Can 2

I • = Mg 0 = 11 += Mt 0= Tr += Wp D= Kp L.=Kn X= Mh T= Pt I A: titanomagnetite 2 0 11 Mt Tr Wp Kp Kn Mh Pt Mg 237.1 256.5 162.7 166.7 79.4 72.3 82.0 68.1 11 11.7 28.9 20.3 76.8 78.6 122.5 140.0 Mt 14.4 14.1 65.8 70.0 108.1 127.1 Tr 6.4 21.5 22.8 50.9 64.6 Wp 36.4 31.9 76.8 89.9 Kp 4.0 10.0 13.6 Kn 16.6 21.3 Mh 2.2

B: hornblende 2 0 11 Mt Tr Wp Kp Kn Mh Pt Mg 37.9 21.1 27.8 27.3 27.6 35.0 15.1 27.6 11 5.6 4.0 9.5 8.0 9.8 21 .6 4.4 Mt 1.4 1.9 4.7 5.8 8.2 3.1 Tr 3.0 5.9 4.8 10.7 2.2 Wp 9.9 8.3 7.8 5.7 Kp 6.6 16.0 6.2 Kn 15.5 5.7 Mh 12.9

Figure 3.4 Plots of the first and second canonical variates (Can1 and Can 2) for Egmont-sourced tephras together with 02 values between tephra groups (tephra abbreviations from Table 3.3). (A) Using titanomagnetite data, (B) using hornblende data.

54 3.3.6 Summaryand Conclusions

Both phenocryst phases examined in this study can be used with the Canonical DFA method to easily and accurately discriminate tephras from EV and TgVC. Titanomagnetite is preferable for this distinction because: (1) there are greater differences in titanomagnetite chemistry between sources, (2) it occurs within nearly all samples, (3) it is relatively stable during weathering, (4) it is rapidly and easily separated from a bulk sample. Canonical OFA of the mean tephra analyses for both of the mineral phases enabled better classification efficiencies of the two volcanic sources and correspondingly larger 02 values, which is consistent with the findings of Shane and Froggatt (1994 ). To apply this technique to discriminating tephras in soil and loess we prefer however, to use a discrimination model based on all of the individual analyses. The analysis of each individual grain of an unknown sample can then be classified by the discriminant model and mixed tephra populations can easily be identified and accounted for. Using this canonical OFA method the probability of any given sample belonging to either of the volcano source groups can be assessed. In situations where unknown samples are close to or within the region of overlap between source groups the probability of group membership is important and can be reported with any correlation made. For the discrimination of individual analyses, titanomagnetite appears very promising. Using the limited database available we can show good discrimination of several individual tephra units, or at least pairs of the stratigraphically closest units, with mostly high 02 values and classification efficiencies. Hornblende provided very little discrimination between tephra groups due to the large spread of data for each tephra as well as small differences in chemistry between tephra groups; only one tephra was clearly discriminated. The most easily distinguished tephra unit, the Manganui tephra was erupted from the parasitic cone of Fanthams Peak on the side of Egmont volcano, whereas the others were probably erupted from the main vent region (AIIoway et al., 1995). This tephra unit is also easily discriminated by its characteristic field appearance and mineralogy. The difference in discriminating abilities of titanomagnetite and hornblende may be related to the petrology of the Egmont volcano system. Stewart et al. (in press) describe titanomagnetite as an early crystallising and stable phase in EV magmas, while hornblende was formed when the melts reached the base of the crust, and on subsequent rising the hornblende was partially resorbed. Stewart et al. (in press) also suggest that these hornblende phenocrysts are largely xenocrysts entrained from lower crusta! and upper mantle sources. This may explain the large spread in the hornblende data for each tephra unit as well as the difficulty in separating the individual tephras. The results of this study demonstrate the potential of the methods described for future tephra discrimination studies of EV and TgVC units as well as providing a more rigorous method of discriminating tephras from the two sources in distal areas. As well as the advantages of OFA for tephra identification reported by others, (Stokes et al.,

55 1992; Shane and Froggatt, 1994) this method is ideal for tephra layers which may be mixed within soil, loess or marine sequences. Using individual grain analyses and their probabilities of group membership, mixed tephra populations can be easily discriminated.

3.3. 7 Acknowledgements

We wish to thank David Lowe for the use of his hornblende analyses, and Ken Palmer (Victoria University of Wellington) for his introduction to the electron microprobe. Thanks are also extended to R.B. Stewart and A.S. Palmer for helpful criticism of an earlier version of this manuscript, as well as P. Froggatt and an anonymous reviewer for their useful comments.

3.4 A MULTIPLE-PARAMETER APPROACH TO ANDESITIC TEPHRA CORRELATION, RUAPEHU VOLCANO, NEW ZEALAND

S. J. Cronin, V. E. Neall, R. B. Stewart, A. S. Palmer Departmentof Soil Science, Massey University, Private Bag 11 222, Palmerston North, New Zealand

1996, Journal of Volcanology and Geothermal Research, 72: 199-215.

Abstract

A multi-parameter approach was used to correlate andesitic tephras in a complex tephra sequence ranging in age from ea. 23 ka to ea. 75 ka on the eastern ring plain of Ruapehu volcano, North Island. Field properties, combined with ferromagnesian mineral assemblages and mineral compositions, were required to map and correlate this sequence. Three tephra units could be identified based on their unique physical appearance, but other tephras could not be correlated on this basis alone. Hornblende and olivine proved to be valuable marker minerals enabling further distinction of two of the marker units recognised by field properties, as well as defining two further marker tephras. Unweathered titanomagnetite crystals, present in all of the tephras, were subjected to major element analysis by electron microprobe. Canonical discriminant function analysis (DFA) of these analyses enabled the grouping and discrimination of tephra units, further aiding the identification of defined marker units, as well as defining new marker units. The titanomagnetite chemistry showed a strong relationship to the ferromagnesian mineralogy, showing that the ferromagnesian phenocrysts formed from the same melt or under the same melt conditions prior to eruption of each tephra. Canonical DFA was also applied to hornblende and olivine mineral analyses to identify further marker beds and to confirm identifications of previously defined units. This statistical analysis was fo und to be invaluable in reducing the large amount of

56 compositional data from this study into a useable form for andesitic tephra correlation and mapping.

3.4.1 Introduction

Andesitic tephrochronology plays an important role in the relative and numerical dating of sediments and geomorphic surfaces on the ring plains of composite volcanoes and surrounding areas. The techniques for andesitic tephra correlation in New Zealand are however less well developed than those for the rhyolitic tephra record of the central North Island. In previous studies of the ring plains of Ruapehu and Tongariro, tephrochronology has played an important role in assigning numerical and relative ages to ring plain surfaces and sediments (Topping, 1973; Topping and Kohn, 1973; Donoghue et a/. , 1995). However studies have been restricted mostly to the younger parts of the ring plain record (< ea. 23 ka.) in which distal rhyolitic tephras present can be correlated to the well established record of the Taupo and Okataina volcanic centres (Froggatt and Lowe, 1990; Wilson, 1993). In this study, ring plain deposits ranging from ea. 23 ka. up to 75 ka are investigated. Radiocarbon dating is less useful in this time frame and fewer rhyolitic tephras are present, with ages less well constrained than those of the post 23 ka record. This constraint places more emphasis on the role of andesitic tephras as local marker horizons for relative dating of surfaces and sediments. In this study we have used multiple criteria to correlate stratigraphic successions and surfaces in different parts of the ring plain. In particular to extend from physical tephra characteristics and mineralogy, canonical discriminant function analysis of titanomagnetite, olivine and hornblende phenocryst compositions were used to correlate andesitic tephras.

3.4.2 Setting

Ruapehu volcano within the Tongariro Volcanic Centre, is a large, active, andesitic stratovolcano, and the highest point in the North Island at 2797 m. Its latest eruption was in 1995. The current massif comprises a 110 km3 cone surrounded by a volcaniclastic ring plain of similar volume (Hackett and Houghton, 1989). The adjacent Tongariro volcano is a slightly smaller andesitic massif made up of several coalescing volcanic cones (Mathews, 1967), the largest of which is the recently active cone of Ngauruhoe. The ring plains of Ruapehu and Tongariro volcanoes are confined to the east by the Kaimanawa Mountains, with the boundary marked by the Tongariro River. The dominant wind direction is from the west, thus the andesitic tephras are generally thickest on the ring plain to the east of the volcanoes. The principal reference sections used in this study are exposed along Upper Waikato Stream (Fig. 3.5), on the north-eastern Ruapehu ring plain.

57 175° 45'

Mt. To ngariro ...

Mt. Ngauruhoe ...

39° 10 I

Mt. Ruapehu ...

0 5km - State Highways SH 1

Figure 3.5. Location map of the eastern ring plains of To ngariro and Ruapehu volcanoes. Numbered dots indicate locations of principal reference sections, numbers correspond to section descriptions contained within Appendix 4.

58 This stream is immediately north of the catchment boundary where present drainage from the eastern flanks of Ruapehu is split between the Whangaehu River to the south and the Tongariro River system to the north.

3.4.3 Methods

A composite stratigraphy of the principal reference sections was first established, providing the longest and most complete record of deposition in the area. The andesitic tephras in this master section were then characterised according to their physical appearance and mineralogy. The simplest and most valuable stratigraphic markers were those which possessed a unique physical appearance enabling their recognition in other areas. The next most valuable markers were those that possessed a unique mineralogy enabling their discrimination from other tephras in the sequence. Those andesitic tephras that could be reliably identified in other sections based on these criteria were few and further fingerprinting was deemed necessary. The major element composition of various mineral phases and andesitic glass were then determined using an electron microprobe for as many of the andesitic tephras as possible. The largest pumice clasts of each tephra sample were cleaned in water with an ultrasonic probe, crushed and their ferromagnesian minerals separated using a Frantz lsodynamic Separator. The mineral grains were then mounted in an epoxy resin plug which was cut and polished (Froggatt and Gosson, 1982). The phases targeted for analysis were titanomagnetite, olivine and hornblende; some analyses were also obtained from andesitic glass, orthopyroxene and clinopyroxene. Orthopyroxene, clinopyroxene, and plagioclase were not used because these minerals all displayed strong oscillatory zoning and greater compositional variation than titanomagnetite, olivine and hornblende. Andesitic glass was poorly preserved in these samples, being present only in some of the younger tephras and in tephras preserved within lignite. Analyses were carried out on a JEOL JXA-733 electron microprobe, employing an accelerating voltage of 15 kV and a 12 nA beam current. A beam diameter of 3 11mwas used for the mineral phases and 1 0 11m with a 8 nA beam current for the andesitic glass (Froggatt, 1983). Mineral standards of the Victoria University of Wellington collection were analysed routinely to correct any machine drift. Between six and ten analyses of individual titanomagnetite crystals were obtained where possible from each tephra sample. Six to ten analyses of hornblende, olivine and andesitic glass were obtained from the samples containing these phases. The resulting data were then examined using traditional oxide plot techniques and statistical analysis.

3.4.4 Statistical methods

Canonical discriminant function analyses were used to establish groups of samples which could be used for correlation and stratigraphic purposes. Canonical discriminant function

59 analysis (OFA) is a technique related to principal component analysis, which reduces the dimensionality of data such as compositional data that has scores on a large number of independent variables. Canonical OFA enables the derivation of a small number of linear combinations of the quantitative variables which best discriminate pre-defined groups of observations or analyses. A reference set of observations or analyses with pre-defined groupings must first be set up and canonical OFA is used to produce a discriminant model which can be used to classify unknown observations or analyses. The theory of OFA is outlined in many texts on multivariate analysis such as Srivastava and Carter (1983) and Johnson and Wichern (1992). The 02 or Mahalanobis distance statistic is produced within the results of canonical OFA, and indicates the multivariate spacing between data groups in multi-dimensions (Srivastava and Carter, 1983). The 02 statistic is then a useful statistic measure of the separation of groups of samples (Stokes et al., 1992), in a manner similar to the numerical coefficients of variation and similarity coefficients of Borchardt et al. (1971 ). OFA was first used to discriminate tephras by Borchardt et al. (1971 ), based on instrumental neutron activation analysis of glasses. King et al.(1982) and Beaudoin and King (1986) used this type of OFA on titanomagnetite major element chemistry to identify Holocene-aged tephras in western Canada. In New Zealand, Stokes and Lowe (1988) and Stokes et al. (1992) demonstrated the use of canonical OFA to discriminate volcanic source areas and later some individual eruptive units, using major element glass analyses. A similar study but including trace and rare earth element analyses was reported by Shane and Froggatt ( 1994 ). These OFA studies also introduced the use of methods which selected variables that had the greatest discriminating power. This method is known as stepwise DFA and is further described by Srivastava and Carter (1983), Stokes and Lowe (1988) and Stokes et al. (1992). Cluster analysis, another form of multivariate analysis, involves the placement of observations into groups suggested by the data only, without any prior groups being established (Johnson and Wichern, 1992). A form of cluster analysis was used by King et al. (1982) and was shown to be ineffe ctive in obtaining a clear, objective discrimination of groups. lt was, however, useful as a preliminary technique to examine any major groupings in the data prior to the use of canonical OF A. In this study the SAS system programs CLUSTER, STEPOISC and CANDISC were used (SAS Institute Inc., 1985). These programs perform cluster analysis, stepwise OFA and canonical OFA, respectively. Prior to statistical analysis the compositional data used in this study was pre treated to avoid statistical closure, in the manner described by Aitchison (1983, 1986) and Stokes and Lowe (1988). The pre-treatment used is termed a log ratio transformation, whereby one oxide score is used to divide into all of the other scores and the logarithm (base 1 0) is taken of each of these ratios. The oxide chosen as the divisor for all of the analyses in this study was MnO, for its moderate abundance and relatively low pooled and within sample variance (Stokes and Lowe, 1988; Shane and Froggatt, 1994).

60 3.4.5 Stratigraphy and distribution of 23-75 ka andesitic tephras

The age of the andesitic tephras in this study is constrained at the top by the Kawakawa Tephra, dated at 22 590 ± 230 years B.P. (Wilson et al. , 1988). The age of the base of the tephra sequence is less well defined and is marked by a large thickness of lahar and stream flow deposits correlated to marine 8180 stage 4 (65-75 ka; Cronin et al., 1996). Throughout the sequence, however, additional dated rhyolitic tephras occur which lend further time control. These rhyolitic tephras are found as glass shards and mineral grains scattered over several centimetres in a matrix of fine-grained andesitic ash. Their identification has been achieved using a combination of electron microprobe major element glass shard chemistry, ferromagnesian mineral assemblages and stratigraphic position (Cronin et al., in press). The microscopic tephras are identified as the Taupe Volcanic Centre-sourced Okaia Tephra and the Okataina Volcanic Centre-sourced Omataroa Tephra, Hauparu Tephra and Rotoehu Ash. The ages of these units are given in Table 3.5, and their relative stratigraphic positions in Fig. 3.6.

Table 3.5 Rhyolitic tephras identified within the eastern ring plain sequence, Ruapehu. Tephra name Sourcea Age (Yrs BP) Reference for age

Kawakawa Tephra TVC 22 590 2301 Wilson et al. (1988) Okaia Tephra TVC ea. 23 0002± Froggatt and Lowe (1990) Omataroa Tephra ovc 28 220 6301 Froggatt and Lowe (1990) ± Hauparu Tephra ovc 35 870 12701 Froggatt and Lowe (1990) ± Rotoehu Ash ovc 64 000 40003 Wilson et al. (1992) ± a TVC = Taupo Volcanic Centre; OVC = Okataina Volcanic Centre. 1 Denotes 14C ages on old half life basis. 2 Estimated stratigraphic age. 3 Whole rock K-Ar age of enclosing lavas.

Andesitic tephras occur throughout the entire sequence at Upper Waikato Stream (Fig. 3.6). The tephras occur as individual pumice lapilli units or as accumulations of fine ash representing the products of several eruptions over time. The lapilli and coarse ash units were used for correlation and fingerprinting studies. Over 60 individual andesitic lapilli units are interbedded within fine ash as well as fluvial and lahar deposits. The distribution of these tephras is not well delimited because their exposure in other sectors of the ring plain is sporadic. They are generally distributed to the east of Ruapehu Volcano and thin rapidly away from their axes of dispersal. Distal equivalents of some of these andesitic lapilli units are mapped to the south in the Rangitikei River valley, lying stratigraphically above the Porewan loess as middle Tongariro Subgroup tephras (Leamy et al. 1973; Milne and Smalley, 1979). Andesitic ash on top of a Porewan loess correlative further to the east in Hawkes Bay is also commonly found (A.P. Hammond, pers. comm., 1995). Partial isopachs of two of the marker horizons are shown in Fig. 3.7. The limited distribution info rmation of these tephras would suggest a Ruapehu volcano source for most of the lapilli units rather than Tongariro volcano.

61 A B c (m) (m) (m) Section Key 20 - �· ��J· ��- Okaia Te phra 0 Kawakawa Te phra /mmMOmatar oa Te phra 0 I � marker unit 5 *I - A 2 0 1 ...... 3 0- B

..... 4 0I � marker unit 1 en 8ft$iJ N ...... - c 5 WJWj D 1

- 30- Rotoehu Ash 11111111111 - Lignite � Sandy matrix debris -.�-'{,�:;� flow deposits Hyperconcentrated [�:�,�;-�] · streamflow deposits G:;'b Silty matrix debris �tq::::: � flow deposits 20--kl� Fine andesitic ash . ·· ··- :I marker unit 3 marker unit 6 • marker unit 7 Andesitic lapilli layers 35 i Strongest paleosol development Distinguishable from rkerunit 4 Bedded sands, silts • physical characteristics / 1:·:·:·:·:·:·1 and gravels I L Figure 3.6 Composite stratigraphic sequence from principal reference sections observed in the Upper Waikato Stream area with positions of andesitic and rhyolitic marker tephras shown. 3.4.6 Physical properties and mineralogy of 23-75 ka andesitic tephras

The coarse-grained andesitic tephra are dominantly fine-medium pumice lapilli which are highly vesicular with very fine vesicles; the proportion of lithic components is usually <1 0 % by volume. The pumice is soft, highly weathered, and ranges from yellow to reddish brown. The degree of weathering, colour, and hardness of the pumice lapilli changes markedly from site to site, depending largely on individual site hydrology. In a single outcrop an individual tephra may range from a pale yellow soft pumiceous lapilli to a reddish brown very firm pumiceous lapilli unit. This makes these parameters virtually unusable for discriminating and correlating the tephra units from place to place even on a very large scale with sites as little as 100 m apart. Shower bedding and graded bedding are also relatively common features and so can only be used in combination with other means to correlate units. Three units can be uniquely identified from the other tephras in the sequence by their physical properties alone (other marker units can additionally be identified using compositional and mineralogy data). The properties of the three units at the principal reference sections are given below, and their stratigraphic positions are shown in Fig. 3.6. Unit 1 is an approximately 400-500 mm thick grey and olive brown lithic coarse ash mixed with yellow and yellow-brown coarse pumice ash. lt is well shower-bedded with alternating 50 mm beds dominated either by grey lithic ash or yellow pumice ash giving a distinctive striped appearance. The basal 20 mm is composed of yellow fine pumice lapilli. The Kawakawa and Okaia Tephras overlie and underlie this unit, respectively, giving a stratigraphic age of ea. 23 ka. The only other tephras found in this area that have a similar appearance to this are members of the Mangamate Tephra described by Topping (1973). The Mangamate Tephra is easily distinguished by stratigraphic position, having been erupted ca.1 0 ka from Tongariro volcano. Unit 2 is a grey, fine ash tuff, which is very hard and breaks up into blocks. lt is showerbedded on a 10 mm scale from very fine to medium ash and contains abundant accretionary lapilli ranging in diameter from 1 to 5 mm within three ea. 40 mm layers. The unit totals ea. 200 mm in thickness. The Omataroa Tephra overlies this unit within a paleosol and the Hauparu Tephra underlies it, giving a stratigraphic age range of 28-36 ka. No other tephra fo und in the study area bears any resemblance to this unit. Unit 3 is a reddish brown and strong brown fine pumice lapilli and coarse ash mixed with grey lithic coarse ash. Two distinctive 10 mm thick bands of grey lithic dominated coarse ash 20 mm apart occur near the top of the unit. This unit is ea. 300 mm thick. The Rotoehu Ash occurs above the unit giving a minimum stratigraphic age of 64 ka. Apart from these three units, there are no other tephra units which can be positively correlated based on their physical features alone. Marker Units 4 and 5 are discriminated on the basis of hornblende and olivine chemistry respectively, while Units 6 and 7 are discriminated using a combination of mineralogy and titanomagnetite chemistry.

63 3.4.7 Mineralogy

The andesitic tephras commonly contain phenocrysts of plagioclase, orthopyroxene, clinopyroxene and titanomagnetite. Some units also contain hornblende and olivine phenocrysts with rare illmenite and chrome spine!. The dominant ferromagnesian assemblage is: 1. orthopyroxene + clinopyroxene > titanomagnetite olivine ± hornblende. ± This assemblage accounts for 90 % of the tephras in this sequence. Hornblende and olivine occur only in small quantities (<2% by volume) in most of the lapilli. Two other ferromagnesian mineral assemblages are also observed in the tephra sequence, and can be characterised as hornblende-dominant and olivine-dominant: 2. Olivine > orthopyroxene + clinopyroxene > titanomagnetite. 3. Hornblende > orthopyroxene + clinopyroxene > titanomagnetite. These assemblages are restricted in their occurrence within the sequence and hence can be used to correlate individual units or small sequences of units to provide further marker beds for stratigraphic study. The bulk of the tephras fit the "type 1" petrological classification of Ruapehu lavas (Cole et al. , 1986; Graham and Hackett, 1987), although much less hornblende occurs in the lavas. The olivine-dominant assemblages fit into the "type 5" lava classification. The lavas with "type 5" mineralogy are generally found in satellite vents rather than on the main Ruapehu cone from where the olivine-dominant tephra appears to have been erupted (Marker unit 1, Fig. 3. 7 A). No hornblende-dominant mineral assemblages are recognised in Ruapehu lavas but rare hornblende-bearing lavas are described from Tongariro volcano (Cole, 1978). Only one tephra has a ferromagnesian mineral assemblage dominated by olivine, (55 % by volume) namely Unit 1 as described previously. This distinctive mineralogy adds to the versatility of this marker bed because it can be identified from its mineralogy in localities where its distinctive physical features may not be evident, e.g. in distal areas or off the dispersal axis. The tephra directly above Unit 1 also contains appreciable olivine (ea. 20% by volume), but all other tephras throughout the entire sequence contain mostly 0-2% by volume olivine. Abundant olivine has only been described previously in Tongariro volcano-sourced members of the Mangamate Tephra (Lowe, 1988a; Donoghue et a/., 1991). Hornblende is another valuable marker mineral because it occurs in only 11 of the tephras in the sequence and mostly in very minor quantities (<1% by volume). lt does, however, occur in greater quantities in two tephras at the very base of the sequence. In these tephras the hornblende content is >5% and in one is up to 30% by volume. The units do not have diagnostic physical features, but they can be used as marker horizons because of their distinctive hornblende rich-mineralogy. The stratigraphic positions of these two tephras (Units 6 and 7) are shown in Fig. 3.6. Both tephras are medium-coarse grade ash and are interbedded within the lignite near the base of the principal reference section.

64 A. B. • • Mt. To ngariro Mt. To ngariro

• • 39° 1 0' Mt. Ngauruhoe Mt. Ngauruhoe

Mt. Ruapehu Mt. Ruapehu • •

Figure 3.7. Partial andesitic tephra isopach maps, thickness given in millimetres. (A) Marker Unit 1, (B) Marker Unit 3.

+ 20

�0 N + O + I= 10 :1: tt+ +++ +

0 I I I I 30 35 40 45 50 FeO % (recalculated)

Figure 3.8. Plot of Ti02 wt % vs. FeO(recalculated) wt % of Titanomagnetite sample data.

65 The Rotoehu Ash occurs stratigraphically higher in the sequence giving a minimum age of 64 ka for Units 6 and 7. Hornblende is also found in small quantities within the previously described marker Unit 3, further enabling its identification as a marker horizon. Hornblende has been previously described only in Tongariro-sourced members of the Mangamate Tephra (Donoghue et al., 1991). Although units 4 and 5 intervene stratigraphically they are distinguished on the basis of their hornblende and olivine chemistry rather than mineralogy and are discussed later in the text.

3.4.8 Mineral chemistry of 23-75 ka andesitic tephras

Table 3.6 Mean electron microprobe analyses of final titanomagnetite, hornblende and olivine groupings. Titanomagnetite Group 1 Group 2 Group 3 Group 4

Si02 0.1 1 (0.08) 0.27 (0.1 0) 0.19 (0.10) 0.15 (0.05) Ti02 21.76 (3.95) 8.91 (2.32) 10.19 (2.28) 10.98 (2.21) Al203 1.09 (0.27) 1.87 (0.32) 3.75 (0.80) 4.49 (1.01 ) FeO 74.78 (5.04) 81.39 (2.27) 81.85 (2.22) 79.78 (2.25) M nO 0.30 (0.07) 0.42 (0.01) 0.41 (0.08) 0.39 (0.08) M gO 1.68 (0.37) 1.69 (0.66) 2.82 (0.65) 3.40 (0.64) Cr203 0.07 (0.03) 0.20 (0.13) 0.42 (0.21) 0.46 (0.21) n 12 12 48 234

Hornblende Group hb1 Group hb2 Group hb3

Si02 41.92 (1.23) 42.20 (0.88) 43.81 (0.30) Ti02 2.46 (0.53) 2.57 (0.60) 1.25 (0.03) Al203 12.93 (1.26) 13.20 (0.94) 12.58 (0.28) FeO 12.64 (1.14) 12.31 (0.61) 14.55 (0.36) M nO 0.30 (0.13) 0.32 (0.11) 0.35 (0.14) M gO 13.94 (0.92) 14.63 (0.65) 14.05 (0.22) CaO 12.32 (0.61) 11.60 (0.59) 11.05 (0.21 ) Na20 2.53 (0.22) 2.49 (0.13) 2.03 (0.05) K20 0.85 (0.15) 0.61 (0.27) 0.28 (0.06) Cl 0.11 (0.11) 0.08 (0.08) 0.05 (0.01) n 30 31 7

Olivine Group ol1 Group ol2 Group ol3 Group ol4

Si02 38.87 (0.88) 36.91 (0.14) 38.51 (1.14) 36.12 (0.28) Ti02 0.04 (0.02) 0.12 (0.01) 0.04 (0.03) 0.02 (0.01 ) AIP3 0.03 (0.01 ) 0.35 (0.17) 0.06 (0.02) 0.07 (0.04) FeO 14.39 (3.94) 31.45 (1.30) 18.11 (3.52) 24.44 (0.25) M nO 0.24 (0.08) 0.49 (0.05) 0.34 (0.15) 0.44 (0.01 ) M gO 46.20 (3.48) 30.13 (1.57) 42.64 (2.78) 38.65 (0.60) CaO 0.14 (0.02) 0.34 (0.04) 0.14 (0.03) 0.08 (0.02) Na20 0.03 (0.03) 0.13 (0.03) 0.05 (0.03) 0.10 (0.05) K20 0.03 (0.02) 0.04 (0.01) 0.07 (0.07) 0.04 (0.01) Cl 0.02 (0.01) 0.03 (0.01) 0.03 (0.02) 0.04 (0.02) n 53 12 46 8

Analyses obtained using a 12 nA beam current at 15 kV, with a 3 1-1mbeam . Standard deviation in brackets.

66 All olivine phenocrysts analysed from tephras studied are forsteritic, ranging in

composition from fo74 to fo87 (Table 3.6). The olivine phenocrysts are not zoned and do not display skeletal morphologies as described by Donoghue et al. (1991 ). Hornblende compositions range from edenite to pargasite (Table 3.6). Clinopyroxene compositions fa ll in the augite field and orthopyroxenes all lie within the enstatite field. Titanomagnetite occurs in a range of ulvospinel compositions (Table 3.6). The mineral chemistry was used to identify any fu rther marker horizons that could be used for stratigraphic study.

3.4.9 Titanomagnetite (discrimination of marker Units 1, 2 and 6)

Titanomagnetite chemistry has been used in several studies to correlate tephra marker beds. In New Zealand, Kahn (1970) demonstrated the use of titanomagnetite composition to distinguish rhyolitic tephras. Kahn and Neall (1973) were able to group andesitic Egmont volcano (EV) tephras into chemically distinct groups and correlate distal andesitic tephras. Distal Tongariro Volcanic Centre-(TgVC) and EV-sourced tephras can also be separated on the basis of their chemistry (Kahn and Neall, 1973; Lowe, 1988a). All of these methods relied on bivariate plots to display differences in selected elements and thus only a small subset of the available composition information is used. King et al. (1982) and Beaudoin and King (1986) used DFA on titanomagnetite compositions to separate individual distal tephra marker beds in Canada. In this study 306 titanomagnetite analyses were made on 53 tephra samples. As a first approach for classification, a number of two oxide plots were examined. The only useful plot for grouping the analyses into coherent groups was the plot of Ti02 vs. FeO (recalculated) (Fig. 3.8). FeO recalculation was from stoichiometry (Droop, 1987). This plot indicates a tightly clustered group of low Ti02 and FeO (recalculated) content samples and a group with higher contents of both oxides; two samples are outliers from these groups. The two outlying samples are those from the marker units 1 and 6. The titanomagnetite oxide analyses were averaged for each sample before being entered into a SAS datafile. The transformed oxides used in the following statistical analyses were Si02, Ti02, Al203, FeO (total as determined by microprobe), MgO, and Cr203. The first approach was assess any 'natural' grouping in the data by using the SAS program CLUSTER (SAS Institute Inc., 1985). The cluster analysis resulted in the 2 recognition of fo ur groups. The Mahalanobis statistics (0 ) between the groups are large, 2 with the closest groups having a 0 value of 8 between them (Fig. 3.9, Table 3.7). The membership of these groups was examined and a relationship between the ferromagnesian mineral assemblage and the clusters was found. Cluster Group 1, contains only one sample, the hornblende-rich marker Unit 6. This unit is well separated from the others in the cluster analysis. Cluster Group 3 contains six samples, five of which are hornblende-bearing. Cluster Group 2 contains ten samples, eight of which contain olivine. Cluster Group 4 contains the bulk of the samples which have mineral assemblages mostly without appreciable olivine or hornblende. There is, however, mixing of samples which have mineral assemblages that do not conform to the group in which

67 the clustering process has placed them. Clustering techniques are usually only an exploratory technique to examine multivariate data (SAS Institute Inc., 1985). The grouping of the samples on the basis of their ferromagnesian mineral assemblage as suggested by the clustering was then tested further. All of the samples were assigned a group according to their ferromagnesian mineral assemblage. These groups were tested using canonical DFA with all the log ratio-transformed oxide scores. The SAS program CANDISC was used (SAS Institute Inc., 1985). The groups were compiled as follows: Group 1 hornblende-rich samples (>5%) Group 2 olivine-rich samples {>10%) Group 3 hornblende-bearing samples {>0.5%) Group 4 all other samples. 2 All of these groups were well discriminated by canonical DFA with high 0 values ranging from 158 to 10 between groups (Fig. 3.98, Table 3. 7). The marker units which are particularly well separated by this analysis are Unit 2 and Unit 6 (Group 1) and Unit 1 (Group 2). Samples of the hornblende bearing tephras in the sequence can also be distinguished from the rest, which may also help to narrow down tephra correlations. If an unknown sample is analysed, this statistical analysis should result in a more precise identification of its possible correlatives. The mean compositions of the titanomagnetites in these four groups are presented in Table 3.6. Stepwise DFA was used to select a subset of variables which enabled the greatest separation between groups. The SAS program STEPDISC was used (SAS Institute Inc., 1985). The variables Ti02, Cr203, MgO, FeO and Al203 were chosen in order as the most discriminating variables. Using these variables in the DFA produced a group separation almost identical to using all of the variables (Table 3.7). Cr203 was chosen as a highly discriminating variable in the analysis but the contents of this oxide is close to the microprobe detection limits in many of the tephras. With Cr203 removed from the analysis good group separation is still produced (Fig. 3.9C, Table 3.7). The order of the remaining variables discriminating power becomes, Ti02, Al203, FeO, and MgO. The well defined DFA groupings show that the differences in major element titanomagnetite composition correlates with differences in phenocryst assemblages of the tephras. Most notably, the hornblende bearing and olivine rich tephras have a different titanomagnetite composition to one another and to the rest of the tephras. Aside from the tephra correlation implications as previously mentioned, the statistical groupings therefore also provide petrological information. In these tephras it appears that the phenocryst phases are genetically related, if not crystallising from the same melt at least the same melt composition for each tephra. This relationship and the fact that all of the tephras examined have a single population of titanomagnetite compositions indicates little magma mixing occurring in this suite of tephras contrasting with a younger Ruapehu-sourced, silica-rich tephra described by Donoghue et al. (1995). The groupings of titanomagnetite samples are not just simply related to the ferromagnesian mineralogy. Other factors appear to affect the group membership. The

68 Ti02 vs. FeO(recalculated) plot described previously shows two main groupings of samples with two outlying samples. The groupings suggested by these two elements were tested using the CANOISC program: Group H High Ti02 and FeO(recalculated) content Group L Low Ti02 and FeO(recalculated) content Group 1 as defined previously (Hornblende rich samples) Group 2 as defined previously (Oiivine rich samples). This procedure clearly separated the groups defined on this basis (Fig. 3.90, Table 3.7). These groupings, which are characterised by different relative contents of Ti02 and FeO(recalculated), could be indicating different eruptive sources e.g. Tongariro vs. Ruapehu, or different batches of magma over time. There appears to be no clear division of Tongariro-sourced and Ruapehu-sourced tephras between Groups H and L. The major element titanomagnetite chemistrydoes not appear to differ consistently for the two volcanoes for this set of samples. Investigating the order of eruption of the tephras indicates that the mixing of groups over time is not random. There are two instances in the tephra record where the titanomagnetites are dominantly of group H, otherwise they are group L. Group H samples, characterised by higher Ti02 and FeO contents may indicate the introduction of a more basic melt at these times. This corresponds with greater quantities of olivine present within the group H tephras. In some cases the greater proportions of olivine are replaced by higher relative quantities of orthopyroxene, hornblende or titanomagnetite in the ferromagnesian mineral assemblage. The outlying groups are again the samples from the marker Unit 1 and Units 2 and 6 as described previously.

2 Table 3.7 0 values between titanomagnetite groupings. 2 Group Definition 0 Group 2 Group 3 Group 4

Cluster Grouping Group 1 (marker Unit 6) 177 104 118 (Fig. 3.9A) Group 2 32 11 Group 3 8

Mineralogical Groupings Group 1 (Marker Units 2, 6) 158 46 50 (Fig. 3.98) Group 2 (Marker Unit 1) 82 149 Group 3 10

Mineralogical Groupings Group 1 (Marker Units 2, 6) 158 45 49 with STEPDISC variables Group 2 (Marker Unit 1) 82 148 Group 3 10

Mineralogical Groupings Group 1 (Marker Units 2, 6) 72 26 41 with reduced variables Group 2 (Marker Unit 1) 59 109 (Fig. 3.9C) Group 3 8

Group L Group 2 Group 1

Chemically defined Group H 9 119 95 Groupings (Fig. 3.90) Group L 95 122 Grou 2 174

69 A. Cluster Groupings B. Mineralogical Groupings 8 I 5 I I I + 4 I 0 r - - ± - - N D N D c:: 0 r------c:: I g ro � D ro �- 0 D 0 ~ I -5 I I -4 ...... I + ... I I

-8 I -10 I

- 12 -8 -4 0 4 8 -5 0 5 10 15 Can 1 Can 1

C. Mineralogical Groupings with D. Chemical Groupings 10 reduced variables 5 I I ...... I 0 0 I 0g 1 + 5 0 - - - l - I r- - N N I , c:: c:: . ro ro I • 0 0 o-�� I 0 - - - - ...... f- -5 I 0�I fSJ t I I D DCf!JD I I t -10 -5 I -10 -5 5 -10 -5 0 5 Can 1 6 Can 1

Symbol Key Group 1 (Marker Units 2, 6) 0 Group 4 .A. + Group 2 (Marker Unit 1) 0 Group H D Group 3 • Group L Figure 3.9. Plots of the first two canonical variables (Can 1 and Can2) for titanomagnetite data. (A) Cluster defined groupings. (B) Mineralogical groupings. (C) Mineralogical groupings with reduced variables (Crp3 omitted). (D) Chemically defined groupings (derived from the Ti02 vs. FeO (recalculated) plot).

70 3.4.10 Hornblende (discrimination of marker Unit 4)

Hornblende chemistry has been used in New Zealand tephra studies as a tool to separate eruptives from EV, TgVC and Taupe Volcanic Zone (TVZ) rhyolitic centres (Lowe, 1988b; Froggatt and Rogers, 1990; Eden et al. , 1993). Eden et al. (1993) used a plot of the atomic proportions of Ca vs. Si to separate EV, TgVC, and TVZ-rhyolitic hornblendes. In this study the hornblende chemistry of individual eruptive units within the tephra sequence was examined to attempt to identify marker horizons in the same manner as for titanomagnetite. A plot of the atomic proportions of Ca and Si did not show any separation or grouping of samples, and no other two oxide plots proved to be useful in distinguishing any marker units. Clustering analysis was applied to the transformed hornblende data in the same manner as for titanomagnetite. As there are only 11 hornblende-bearing tephras analysed in this sequence, the mean analyses of each sample cannot be used in the multivariate statistical procedures due to the lack of sufficient degrees of freedom. As a result, individual analyses were used in the following procedures rather than means. The clustering procedure produced three groups of samples with good separation, Groups hb1 , hb2 and hb3 (Fig. 3.1 OA, Table 3.8). Several tephra samples had all of their analyses contained within a single group, but some of the samples had one or two of their analyses in another cluster grouping. The hornblende cluster groups were then refined by keeping the analyses of each tephra sample within a single group (i.e. assuming a single population for each tephra sample). Canonical DFA of the refined hornblende groups produced better discrimination with higher D2 values (Fig. 3.1 OB, Table 3.8), indicating that there are no mixed populations of hornblendes in these samples. The STEPDISC program chose, in order, K20, FeO, Ti02, Al203, MgO, CaO, and Si02 as the most discriminating variables. Using this subset of variables the group separation was almost as good as using all of the variables (Table 3.8). The hornblende groupings enable the discrimination of marker Unit 4 from the other hornblende bearing tephras. The refined Group hb3 is well separated from the other groups and consists of analyses of marker Unit 4 (Fig. 3.6). This unit had not been uniquely identified with any of the methods described previously. Group hb2 contains the analyses of several tephra samples but they all lie within the same part of the sequence, in the middle part of the record below the distinctive, previously described marker Unit 3 (Fig 3.6). Group hb1 contains the tephra samples which are at the very base of the sequence, including marker Unit 6. Group hb1 also contains the distinctive marker Unit 2, but this is able to be identified on the basis of its field appearance. The hornblende major element chemistry can be used to potentially place an unknown tephra sample in a specific part of the sequence, and in the case of one tephra (marker Unit 4) uniquely identify it.

71 4 A. Cluster Grouping 5 A. Cluster Grouping tJ D 0 "* 2 fj;O 1 D - - - 0 1------� 0 f- o+�� 0 or D-£J� D I N N ' c: I roc: ro CJD D � 0 0 0 -2 �*I� + I -5 + •t D I -4 I 0 I I + I

-6 I -10 I ' -4 -2 0 2 4 6 -10 -5 0 5 10 Can 1 Can 1

4 B. Refined Groupings 10 B. Mineralogical Groupings I 0 1 0 o p o oo l + Group ol1 2 0 0 0 o Group ol2 o 40� o 5 I 0 Group ol3 D � o D � �� 0 0 1 0 Group ol4 (Marker Unit 5) N N c: B 0 0 0 - c: ro 0 ------ro 0 - + I +0 0 @0 + I � - - + 0 -+�CO - -G- + I - -2 -!+ - - + J.+ + + -++�l I I D + � �. -4 +----r-----..----,---+-1 -�-...., -5 +---�--I ---.------8 -6 -4 -2 0 2 4 -5 0 5 10 15 Can 1 Can 1

C. Mineralogical Groupings 5 with reduced variables

0 Group hb1 I + Group hb2 + I D Group hb3 (Marker Unit 4)

N - - - roc: 0 1- - -o- 0 - Figure 3.1 O.Piots of the first two canonical variables for hornblende data. (A) Cluster defined 0 l D groupings. (B) Refined cluster-defined o groups. <»�� I I -5 4----+-�--.....---�--� -5 0 5 10 15 Can 1

Figure 3.11. Plots of the first two canonical variables for olivine data. (A) Cluster defined groupings. (B) Mineralogical groupings. (C) Mineralogical groupings with reduced variables.

72 2 Table 3.8 0 values between hornblende groupings.

Group definition 02 Group hb2 Group hb3

Cluster Groupings Group hb1 11 11 (Fig. 3.1 OA) Group hb2 11

Group hb2 Group hb3 {Marker Unit 4} Refined Groupings Group hb1 11 39 (Fig. 3.1 OB) Group hb2 35

Refined Groupings with Group hb1 9 37 STEPDISC variables Group hb2 34

3.4.1 1 Olivine (discrimination of marker Unit 5)

Olivine has been used in New Zealand tephrochronology studies for distinguishing Egmont- and Tongariro-sourced tephras. Lowe (1988a) pointed out the presence of forsteritic olivine within TgVC-sourced tephras but very little olivine within EV-sourced tephras. Oonoghue et al. (1991 ), used the morphology and chemistry of olivine phenocrysts to identify individual tephra members of the Tongariro volcano-sourced Mangamate Tephra Formation. Almost half of the tephras within this sequence contain olivine and approximately half of the olivine-bearing tephras contain more than trace amounts (>0.5%) of olivine phenocrysts. No skeletal forms of olivine analogous to those described by Oonoghue et al. (1991) in the Mangamate Tephra Formation were seen. In this study 119 grains of olivine were analysed from 16 tephras. Bivariate oxide plots showed no consistent groupings of samples that were easily discernible by eye because oxide values are very similar from one analysis to another. Clustering analysis was applied to the transformed olivine data in the same manner as for titanomagnetite. The oxide variables used were Si02, Al203, FeO, MgO, CaO, and Na20, the other oxides reported in Table 3.6 were too low and close to microprobe detection limits. Four groupings resulted, which were separated by significant 2 0 spacings (Fig. 3.11A, Table 3.9). The group membership of tephra units was found to be dominantly (but not exactly) the same as the previously defined titanomagnetite groupings. With this in mind, groups were defined on a similar basis to the titanomagnetite groups for canonical OFA: Group ol1 Olivine rich samples (>5%) Group ol2 Hornblende bearing samples Group ol3 All other samples Group ol4 Marker Unit 5 Group ol4 had to be defined to contain analyses of a tephra sample which did not fit into any of the other groups (marker Unit 5). These fo ur groups were well separated using OFA (Fig. 3.11B; Table 3.9). The STEPOISC procedure chose, in order: MgO, Si02 Al203, CaO, and FeO as the most discriminating variables. Using this subset of variables,

73 group separation almost as good as using all of the variables was achieved (Table 3.9). The DFA was also tried without the STEPDISC chosen oxides Al203 and CaO which are of low content in these olivines. The resultant discrimination is still good for all groups but was reduced between groups ol1 and ol3 (Fig. 3.11C, Table 3.9). Using these statistical techniques with olivine analyses from the tephra sequence the group separation defined by the titanomagnetite chemistry is able to be corroborated. Thus the olivine chemistry appears to be as sensitive to the ferromagnesian assemblage as the titanomagnetite chemistry. One further tephra unit can be uniquely identified in the sequence on the basis of its olivine chemistry. This is the unit represented by Group ol4, corresponding to marker Unit 5 (Fig. 3.6).

Table 3.9 02 values between olivine groupings. 2 Group definition D Group ol2 Group ol3 Group ol4 (Marker Unit 5) ·

Cluster Groupings Group ol1 40 11 89 (Fig. 3.1 1 A) Group ol2 35 113 Group ol3 66

Mineralogical Groupings Group ol1 145 9 46 (Fig. 3.11 B) Group ol2 106 169 Group ol3 22

Mineralogical Groupings Group ol1 142 7 34 with STEPDISC Group ol2 106 166 variables Group ol3 18

Mineralogical Groupings Group ol1 107 1.8 19 with reduced variables Group ol2 93 161 (Fig. 3.11 C) Group ol3 15

3.4.12 Other phases

Lowe (1988a) found some provisional separation between EV and TgVC pyroxenes using analyses from the phenocryst cores. The pyroxene grains of the present sequence, however, were too strongly compositionally zoned to attempt chemical correlation and discrimination between individual eruptives. Andesitic glass chemistry was shown to be effective in discriminating EV and TgVC tephras (Lowe, 1988a; Stokes and Lowe, 1988). The preservation of andesitic glass is, however, poor in most sedimentary environments, and very poor in the New Zealand soil environment (Neall, 1977; Kirkman and McHardy, 1980). Very few glass analyses could be obtained from the andesitic tephras in this sequence, and those that were obtained were considered unreliable due to the degree of weathering and hydration of the glass.

74 3.4.13 Conclusions

1) In this andesitic tephra sequence, ranging in age from ea. 23 ka to 75 ka, correlation of individual units over the eastern Ruapehu ring plain has been achieved through a variety of laboratory and field methods (Table 3.1 0). Seven units can be distinguished as marker beds over the eastern Ruapehu ring plain. This enables relative dating of ring plain deposits and surfaces, where other dating methods are less useful or more expensive.

Table 3.10 Summary of the criteria for identification of andesite marker tephras in this study. Tephra marker Unique field Characteristic Unique Unique Unique appearance mineralogy titanomagnetite hornblende olivine chemist!Y chemist!Y chemist!:l

Unit 1 ./ olivine rich ./

Unit 2 ./ hornblende rich ./

Unit 3 ./ hornblende ./ bearing Unit 4 hornblende ./ bearing Unit 5 ./

Unit 6 hornblende rich ./

Unit 7 hornblende rich

2) Three tephra units within this sequence are able to be identified over the eastern Ruapehu ring plain based on their unique field appearances; no other units can be uniquely identified on this basis. The field-based marker units are termed Units 1, 2, and 3. 3) The ferromagnesian mineral assemblage is another valuable identification criterion. The presence of >50 % modal olivine and > 5% modal hornblende are useful in further characterising Units 1 and 2, respectively. The presence of large quantities of hornblende (>5 to 30% by volume) enables the identification of two further tephra units near the base of the sequence, marker Units 6 and 7. Further identification of marker beds relies on electron microprobe analyses of phenocryst minerals. 4) The use of statistical clustering methods on electron microprobe-determined mineral compositions was found to have limited use in accurately grouping and discriminating mineral data, but was a valuable reconnaissance technique. Canonical DFA was used to distinguish groupings of sample data which were refined from clustering analysis by using other factors such as ferromagnesian mineral assemblage. 5) The grouping and discrimination of tephras using their titanomagnetite major element analyses reflected the tephra ferromagnesian mineral assemblage. Four distinct chemical groups were defined:

75 Group 1 hornblende-rich samples (>5%) Group 2 olivine-rich samples (> 10%) Group 3 hornblende-bearing samples Group 4 all other samples This implies that all of the phenocrysts in these andesitic tephras were formed together in the same crystallising event or under the same magmatic conditions, and that these conditions are also reflected by the titanomagnetite chemistry. Titanomagnetite mineral chemistry was used to further discriminate marker Units 1, 2, and 6, as well as narrow down the identification of the other units. This is especially useful in situations such as distal areas where the ferromagnesian mineral assemblage is not well represented. The titanomagnetite chemistry also indicates episodes of changing melt composition during the eruptive history of Ruapehu volcano. Two episodes, thought to represent a more basic melt composition were seen. 6) Marker Units 2 and 6 are able to be further identified on the basis of hornblende composition and one other unit is able to be uniquely identified: marker Unit 4. The remaining hornblende bearing units are specifically grouped according to their stratigraphic position, enabling correlation to a general part of the sequence if not a specific tephra unit. 7) Olivine chemistry corroborates the groupings of tephra samples based on their titanomagnetite analyses, giving further means of identifying Unit 2 and the groups defined by the titanomagnetites. One other tephra unit can also be uniquely identified by its olivine chemistry: marker Unit 5. 8) In this study the usefulness of a multi-parameter approach and the statistical analysis of data for andesitic tephra correlation has been shown. In an andesitic ring plain environment very few tephra units can be discriminated by physical features alone as the physical appearances of small-medium volume andesitic tephra units change over very small distances. Detailed field mapping needs to be combined with increasingly diverse techniques to be able to correlate such tephras. The described statistical methods have proven useful in conjunction with more traditional methods currently used for andesitic tephra fingerprinting and correlation.

3.4.14 Acknowledgements

SJC gratefully acknowledges funding from the New Zealand Vice-Chancellor's Committee, Massey University Graduate Research Fund, and the Helen E. Akers Scholarship Fund. We thank K. Palmer for introduction to the use of the electron microprobe, and B. Roser and D. Lowe for their thoughtful reviews and comments.

76 3.5 Combined reference list

Aitchison, J., 1983. Principal component analysis of compositional data. Biometrika, 70: 57-65. Aitchison, J., 1986. The Statistical Analysis of Compositional Data. Monographs on Statistics and Applied Probability. Chapman and Hall, London. Alloway, B.V., Neall, V.E., Vucetich, C.G., 1995. Late Quaternary (post 28 000 year B.P.) tephrostratigraphy of northeast and central Taranaki, New Zealand. J. Roy. Soc. N.Z., 25: 385-458. Beaudoin, A.B. and King, RH., 1986. Using discriminant function analysis to identify Holocene tephras based on magnetite composition: a case study from the Sunwapta Pass area, Jasper National Park. Can. J. Earth Sci., 23: 804-812. Borchardt, G.A.., Harward, M.E. and Schmitt, RA., 1971. Correlation of ash deposits by activation analysis of glass separates. Quat. Res., 1: 247-260. Cole, J.W., 1978. Andesites of the Tongariro Volcanic Centre, North Island, New Zealand. J. Volcano!. Geotherm. Res., 3: 121-153. Cole, J.W., Graham, I.J., Hackett, W.R and Houghton, B.F., 1986. Volcanology and petrology of the Quaternary composite volcanoes of Tongariro Volcanic Centre, Taupe Volcanic Zone. In: Smith, I.E.M. Late Cenozoic volcanism in New Zealand. Roy. Soc. N.Z. Bull., 23: 225-250. Cronin, S.J., Neall, V.E. and Palmer, A.S., 1996. The geological history of the north eastern ring plain of Ruapehu Volcano, New Zealand. Quat. Inter., 34-36: 21-28. Donoghue, S.L., Neall, V.E. and Palmer A.S. 1995. Stratigraphy and chronology of late Quaternary andesitic tephra deposits, Tongariro Volcanic Centre, New Zealand. J. Roy. Soc. N.Z., 25: 115-206. Donoghue, S.L., Stewart, RB. and Palmer, A.S., 1991 . Morphology and chemistry of olivine phenocrysts of Mangamate Tephra, Tongariro Volcanic Centre, New Zealand. J. Roy. Soc. N.Z., 21: 225-236. Droop, G.T.R, 1987. A general equation for estimating Fe2+ concentrations in ferromagnesian silicates and oxides from microprobe analyses, using stoichiometric criteria. Min. Mag., 51: 431-435. Eden, D. N., Froggatt, P.C., Trustrum, N.A. and Page, M.J., 1993. A multiple-source Holocene tephra sequence from Lake Tutira, Hawkes Bay, New Zealand. N.Z. J. Geol. Geophys., 36: 233-242. Froggatt, P.C., 1983. Towards a comprehensive Upper Quaternary tephra and ignimbrite stratigraphy in New Zealand using electron microprobe analysis of glass shards. Quat. Res., 19: 188-200. Froggatt, P.C. and Gosson, G.J., 1982. Techniques for the preparation of tephra samples for mineral or chemical analysis and radiometric dating. Geology Department, Victoria University of Wellington Publication 23 (12 pp.).

77 Froggatt, P.C. and Lowe, D.J., 1990. A review of late Quaternary silicic and some other tephra formations from New Zealand: their stratigraphy, nomenclature, distribution, volume, and age. N.Z. J. Geol. Geophys., 33: 89-109. Froggatt, P.C. and Rogers G.M., 1990. Tephrostratigraphy of high-altitude peat bogs along the axial ranges, North Island, New Zealand. N.Z. J. Geol. Geophys., 33: 111-124. Graham, I.J. and Hackett, W.R., 1987. Petrology of calc-alkaline lavas from Ruapehu volcano and related vents, Taupo Volcanic Zone, New Zealand. J. Petrol., 28: 531-567. Hackett, W.R. and Houghton, B.F., 1989. A facies model for a Quaternary andesitic composite volcano: Ruapehu, New Zealand. Bull. Volcano!., 51 : 51-68. Hodder, A.P.W., de Lange, P.J., Lowe, D.J., 1991. Dissolution and depletion of ferromagnesian minerals from Holocene tephra layers in an acid bog, New Zealand, and its implications for tephra correlation. J. Quat. Sci., 6: 195- 208. Johnson, R.A. and Wichern, D.W., 1992. Applied Multivariate Statistical Analysis (3rd Edition). Prentice-Hall Inc. New Jersey. Juvigne, E., Shipley, S., 1983. Distribution of the heavy minerals in the downwind tephra lobe of the May 18, 1980 eruption of the Mount St. Helens (Washington, USA). Eiszeitalter U. Gegenwart 33: 1-7. King, R.H., Kingston, M.S. and Barnett, R.L., 1982. A numerical approach to the classification of magnetites from tephra in southern Alberta. Can. J. Earth Sci., 19: 2012-2019. Kirkman, J.H. and McHardy, W.J., 1980. A comparative study of the morphology, chemical composition and weathering rhyolite and andesite glass. Clay Minerals, 15: 165-173. Kohn, B.P., 1970. Identification of New Zealand tephra layers by emission spectrographic analysis of their titanomagnetites. Lithos, 3: 361 -368. Kohn, B.P. and Neall, V.E., 1973. Identification of late Quaternary tephras for dating Taranaki lahar dep9sits. N.Z. J. Geol. Geophys., 16: 781 -792. Leamy, M.L., Milne, J.D.G., Pullar, W.A. and Bruce, J.G., 1973. Paleopedology and soil stratigraphy in the New Zealand Quaternary succession. N.Z. J. Geol. Geophys., 16: 723-744. Lowe, D.J., 1986. Controls on the rates of weathering and clay mineral genesis in airfall tephras: a review and New Zealand case study. In: Colman, S.M., Dethier, D.P. (eds). Rates of chemical weathering of rocks and minerals. Academic Press, Orlando: 265-330. Lowe, D.J., 1987. Studies on late Quaternary tephras in the Waikato and other regions in northern North Island, New Zealand, based on distal deposits in lake sediments and peats. Unpub. PhD thesis, University of Waikato, Hamilton, New Zealand. Lowe, D.J., 1988a. Stratigraphy, age, composition, and correlation of late Quaternary tephras interbedded with organic sediments in Waikato lakes, North Island, New Zealand. N.Z. J. Geol. Geophys., 31 : 125-165.

78 Lowe, D.J., 1988b. Late Quaternary volcanism in New Zealand: towards an integrated record using distal airfa ll tephras in lakes and bogs. J. Quat. Sci., 3: 111-120. Lowe, D.J., 1990. Tephra studies in New Zealand: an historical review. J. Roy. Soc. N.Z.,20: 119-150. Mathews, W.H., 1967. A contribution to the geology of the Mount Tongariro massif, North Island, New Zealand. N.Z. J. Geol. Geophys., 10: 1027-1038. Milne, J.D.G. and Smalley, I.J., 1979. Loess deposits in the southern part of the North Island of New Zealand: an outline stratigraphy. Acta Geologica Academiae Scientarium Hungaricae, 22: 197-204. Neall, V.E., 1977. Genesis and weathering of Andisols in Taranaki, New Zealand. Soil Sci., 123: 400-408. Pullar, W.A., 1967. Uses of volcanic ash beds in geomorphology. Earth Sci. J., 1: 164- 177. Randle, K., Gorton, G.G., Kittleman, L.R., 1971. Geochemical and petrological characterisation of ash samples from Cascade Range volcanoes. Quat. Res., 1: 261 -282 Ruxton, B.P., 1968. Rates of weathering of Quaternary volcanic ash in north-eastern Papua. 9th International Congress of Soil Science Transactions Vol. 4, Paper 38: 367-376 SAS Institute Inc., 1985. SAS Users guide: Statistics. Version 5 Edition, SAS Institute Inc., Cary N.C., 956 pp. Shane, P.A.R. and Froggatt, P.C., 1994. Discriminant function analysis of glass chemistry of New Zealand and North American tephra deposits. Quat. Res., 41 :70-81. Smith, D.G.W., Westgate, J.A. , 1969. Electron probe technique for characterising pyroclastic deposits. Earth Planet. Sci. Lett., 5: 313-319. Srivastiva, M.S. and Carter, E. M., 1983. An Introduction to Applied Multivariate Statistics. Elsevier Science Publishing Co. Inc., New York. Stewart, R.B., Price, R.C., Smith, I.E.M., (in press). Evolution of high-K arc magma, Egmont volcano, Taranaki, New Zealand: evidence from mineral chemistry. J. Volcano!. Geotherm. Res. Stokes, S. and Lowe, D.J., 1988. Discriminant function analysis of late Quaternary tephras from five volcanoes in New Zealand using glass shard major element chemistry. Quat. Res., 30: 270-283. Stokes, S., Lowe, D.J. and Froggatt, P.C., 1992. Discriminant function analysis and correlation of Late Quaternary rhyolitic tephra deposits from Taupo and Okataina volcanoes, New Zealand, using glass shard major element composition. Quat. Inter., 13/14: 103-1 17. Thorarinsson, S., 1949. Some tephrochronological contributions to the volcanology and glaciology of Iceland. Geografiska Annaler, 31: 239-256.

79 Topping, W.W., 1973. Tephrostratigraphy and chronology of late Quaternary eruptives from the Tongariro Volcanic Centre, New Zealand. N.Z. J. Geol. Geophys., 16: 397-423. Topping, W.W. and Kohn, B.P., 1973. Rhyolitic tephra marker beds in the Tongariro area, North Island, New Zealand. N.Z. J. Geol. Geophys., 16: 375-395. Vucetich, C.G., Pullar, W.A. , 1969. Stratigraphy and chronology of late Pleistocene volcanic ash beds in central North Island, New Zealand. N.Z. J. Geol. Geophys., 12: 784-837. Vucetich, C.G., Pullar, W.A., 1973. Holocene tephra formations erupted in the Taupo area and interbedded tephras from other volcanic sources. N.Z. J. Geol. Geophys. 16: 745-780. Wallace, R.C., Alloway, B.V., Stewart, R.B., Neall, V.E., 1986. Mineral chemistry as an indicator of petrogenisis at Egmont volcano. Abstracts of the International Volcanological Congress, 1986 New Zealand: 23. Wallace, R.C., 1987. The mineralogy of the Tokomaru Silt Loam and the occurrence of cristobalite and tridymite in selected North Island soils. Unpub. Ph.D. thesis, Massey University, New Zealand. Wilson, C.J.N., 1993. Stratigraphy, chronology, styles and dynamics of late Quaternary eruptions from Taupo volcano, New Zealand. Phil. Trans. Roy. Soc. London A, 343: 205-306. Wilson, C.J.N., Houghton, B.F., Lanphere, M.A. and Weaver, S.D., 1992. A new radiometric age estimate for the Rotoehu Ash from Mayor Island volcano, New Zealand. N.Z. J. Geol. Geophys., 35: 371-374. Wilson, C.J.N., Switzur, R.V. and Ward, A.P., 1988. A new 14C age for the Oruanui(Wairakei) eruption, New Zealand. Geol. Mag., 125: 297-300.

80 CHAPTER 4: PALEOSOL DEVELOPMENT AND PALEOCLIMATIC INVESTIGATIONS OF THE RING PLAIN SEQUENCE

4.1 Introduction

In the previous two chapters a stratigraphic framework of rhyolitic and andesitic tephras has been established for the older ring-plain sequences found on the NE Ruapehu and E Tongariro ring plains. The first way in which the stratigraphic framework has been applied was the investigation through time of weathering and soil development within the ring plain sequence. In particular the stratigraphic framework allowed the sequence of weathering and soil development on the ring plains to be correlated to other investigations of the New Zealand late Quaternary paleoclimatic record. In the following published study, parameters of weathering and soil development in the ring plain sequences, particularly the clay mineralogy, were investigated. In this way past climatic and soil development conditions on the ring plain areas were elucidated. The sequence of soil development was compared to other paleosol sequences in other areas of the North Island of New Zealand using the interbedded tephra layers as time planes. The following paper has multiple authors and the contributions of each author to the work were:

S.J. Cronin: Principal investigator Carried out all: field description and sampling laboratory preparation of samples optical microscopy work chemical, XRD and DTA analyses grainsize and soil-physical analyses manuscript preparation and writing

V.E. Neall and A.S. Palmer: Advisers Aided the study by: discussion of methodology and results editing and discussion of the manuscript

81 4.2 INVESTIGATION OF AN AGGRADING PALEOSOL DEVELOPED INTO ANDESITIC RING-PLAIN DEPOSITS, RUAPEHU VOLCANO, NEW ZEALAND.

Shane J. Cronin, Vincent E. Neall, and Alan S. Palmer Department of Soil Science, Massey University, Private Bag 11222, Palmerston North, New Zealand.

Geoderma, 69 (1996) 119-135.

Abstract Within a sequence of andesitic volcaniclastic deposits on the north-eastern ring plain of Ruapehu volcano is a ca.1 0 m-thick sequence of weathered andesitic tephras. Weathering and paleosol development is most evident in 3.6 m of fine ash in this sequence. The age of these tephras are constrained between ea. 23-70 ka by dated rhyolite tephras erupted from central North Island volcanoes. Mineralogy of the fine ash deposits reveals their origin, and the processes involved in their soil development. The fine ash deposits are almost totally locally derived either as primary volcanic ash or fines reworked from the ring plain itself by aeolian processes. Aerosolic quartz input associated with regional loess deposition during the cool climatic episodes of mid-8180 stage 3 and 8180 stage 4 is very low, having been diluted by rapid accumulation of andesitic tephras in these episodes. The observed weathering features and secondary minerals within the ash sequence were derived from a complex combination of factors including climate change, accretion rate, and post-depositional modification. Relatively strong weathering development in two parts of the ash sequence is correlated with two widespread soil development episodes during the Last Glacial observed throughout the southern North Island. The accretion rate of the soil surface at these times also affected the expression of climate-related weathering. Formation of allophane (with an AI/Si ratio of 2:1) and ferrihydrite occurred near the soil surface as the ash was accreting. The amount of allophane and ferrihydrite through the sequence appears to be inversely related to the accretion rate of the soil surface. Upon burial of the ash materials by a thick (>20 m) sequence of lahar and tephra deposits, halloysite was later formed in the buried ash. The leaching of silica from the thick overburden of volcaniclastics into the ash material as well as perched water is thought to have decreased the AI/Si ratio in the soil solution and thus promoted the fo rmation of halloysite from weathering andesitic glass.

4.3 Introduction

Preserved on the eastern ring plain of Ruapehu volcano is a >50 m thick sequence of volcaniclastic deposits which have accumulated over the last ea . 75 ka (Cronin, et al. , 1996). Within this sequence the contrasting morphologies of the deposits indicate the intensity of surficial weathering has varied over time. During the cold climatic episodes of marine 8180 stages 2 and 4, near-continuous aggradation of diamictons occurred over the entire eastern Ruapehu ring plain (diamicton is a non genetic term for poorly sorted

82 sediments which contain a wide variety of particle sizes; Flint, 1960). The diamictons of this sequence generally contain pebble- to boulder-sized clasts which are supported by a sand-silt matrix. The aggradation of the diamictons is thought to be a reflection of unstable conditions on Ruapehu volcano, and the mechanism of deposition is thought to be dominantly by lahars (volcanic mud-flows). Within the periods of rapid aggradation on the ring plain there appears to have been little opportunity for the preservation of any air fall tephra or volcanic loess. The only tephras preserved in the diamicton sequences are the very large lapilli units and these have been reworked and eroded in many places. There is no soil development evident within the "packages" of diamictons. During the mild climatic episodes of 8180 Stage 3, deposition on the ring plain was dominated by tephra and volcanic loess accession. Soil development is evident throughout these deposits but is most strongly exhibited in the fine ash material. Diamictons still occur in the sequence but were separated by significant time intervals. The aim of this study was to characterise the origin of the fine ash deposits and to investigate the paleosol-forming environment throughout the sequence of tephra and volcanic loess deposits. This was achieved through field interpretation of the deposits coupled with mineralogical studies.

4.4 Setting

Ruapehu volcano within the Tongariro Volcanic Centre, is a large, active, andesitic, stratovolcano, and the highest point in the North Island at 2797 m. The current massif comprises a 110 km3 cone with an apron of volcaniclastic deposits surrounding it of similar volume (Hackett and Houghton, 1 989). The apron of volcaniclastic deposits is termed the ring plain. The area of this study is located along the Upper Waikato Stream of the north-eastern ring plain {Chapter 3: Fig. 3.5). Exposed in continuous stream bank exposures in this and adjacent streams are the deposits described in this paper. The altitude of the area ranges from 900 to 1100 m above sea level. The Ruapehu region has a cool-temperate climate with numerous frosts and snow-falls, and a mean annual temperature of 10 oc at 600 m (McGione and Topping, 1977). The orographic influence of Ruapehu volcano on the predominantly westerly airflow increases the rainfall on the higher slopes of the volcano compared to surrounding areas. The average annual rainfall varies from 1600 mm at 900 m to 2400 mm at 1100 m altitude on the eastern side of the volcano (N.Z. Meteorological Service, 1973).

4.5 Stratigraphy

The tephra dominant portion of this section of the ring plain sequence occurs between two "packages" of diamictons (Fig. 4.1 ). Time control is provided by the presence of five rhyolitic tephras erupted from central North Island volcanoes to the north (Table 4.1 ). The four oldest tephras occur as rhyolitic glass shards and minerals dispersed over several centimetres within a matrix of fine andesitic ash. The identification of these tephras was

83 made on the basis of characteristic mineralogy and electron-microprobe major element glass chemistry.

Table 4.1 . Rhyolitic tephra identified within the north-eastern ring plain sequence.

Tephra Name Source Age Reference for age

Kawakawa Tephra TVC 22 590 2301 Wilson et al., (1988) ± Okaia Tephra TVC ea. 23 000 Froggatt and Lowe, (1990) Omataroa Tephra ovc 28 220 6301 Froggatt and Lowe, (1990) ± Hauparu Tephra ovc 35 870 12701 Froggatt and Lowe, (1990) ± 2 Rotoehu Ash ovc 64 000 4000 Wilson et al., (1992) ±

TVC = Taupo Volcanic Centre OVC = Okataina Volcanic Centre 1 Denotes 14C ages based on old half life (years B.P.) 2 Whole rock K-Ar age (years)

4.6 Physical description

The ash materials of this study are 20 to 30 m below the present soil surface. The diamictons interbedded with, and above and below the tephra and volcanic loess sequence have a sandy matrix supporting clasts of andesitic lava; the clasts can be over 1 m in diameter. All of the diamictons contain 5-10% by volume andesitic pumice lapilli but in an extreme case up to 60%. Two of the diamicton units in this sequence are firmly cemented, while the others are unconsolidated. The andesitic tephra occur as individual pumice lapilli and coarse ash beds or as accumulations of fine ash representing the products of several eruptions over time. The coarser andesitic tephra are dominantly fine pumice lapilli which are highly and very finely vesicular; the proportion of lithic components is usually <1 0% by volume. The pumice is soft, highly weathered, and ranges from yellow to reddish-brown. The pumice lapilli occur interbedded with fine ash throughout the sequence. A high frequency of pumice lapilli units within fine ash is here considered to represent more rapid accumulation of the surface compared to slower accretion where pumice lapilli are fewer within the fine ash. The finer grained andesite ash ranges in texture from fine sandy loam to loam. A few zones within the ash column contain the occasional pumice lapilli clasts while most of these deposits are uniformly fine grained <2 mm. The ash is dominantly yellow-brown and brown and some sites have lower chroma colours but little or no evidence of mottling. lt is firm when moist and dries into very strong blocks. Very little soil structure is seen apart from coarse blockiness when the ash material is dry. Root rhizomorphs are abundant throughout all of the fine ash. Two zones of more strongly weathered ash with common to very common root rhizomorphs were noted (Fig. 4.1 ). The upper zone is dark brown, contains finely disseminated charcoal fragments, and has the greatest abundance of root rhizomorphs.

84 Overall Sequence

�+++++++-Om

25 Cemented Diamicton

Diamictons 30 �+++++++- Re � Andesitic · � ;. ·� '6· • "': o-:- :. . ,_; • · · · .. ,: _ ., . · · .. � Lapilli Fine Andesitic D Ash Stronger paleosol [ill]development

Figure 4.1 . Stratigraphic column of part of the northeastern Ruapehu ring-plain sequence with paleosol development and its position within the overall ring plain sequence. Kk =

Kawakawa Te phra, Ok = Okaia Te phra, Om = Omataroa Te phra, Hu = Hauparu Te phra, Re = Rotoehu Ash.

85 The lower zone is not as well expressed but contains a greater abundance of root rhizomorphs than the surrounding ash, as well as finely disseminated charcoal fragments. The dry bulk density of the ash material ranges from 0.68 to 0.86 Mg m·3 (Table 4.2). The lowest bulk densities are in the upper part of the ash column within the upper zone of stronger weathering. The highest bulk densities are within the lower zone of stronger weathering.

Table 4.2. Soil physical properties of selected ash samples.

Sample depth Dry bulk density Field-moist Air-Dry gravimetric Gravimetric water 3 (m) (Mg m· ) gravimetric water water content at content of air dry content at -1 . 5 -1.5 MPa (%) soil <2mm (%) MPa %

0.4-0.5 0.68 78 50 13 0.9-1 .0 0.68 96 52 23 1.5-1 .6 0.86 64 51 15 1.8-1 .9 0.80 81 50 10 3.5-3.6 0.75 75 39 15

4. 7 Mineralogy of ash grade material

Selected mineralogy of the ash grade material was investigated to identify the provenance of the material and to elucidate the soil forming processes occurring in the ash. The fine ash was channel sampled at 10 cm intervals through its 3.6 m thickness.

4.8 Methods

Quartz, cristobalite, and plant opal were concentrated from the whole soil (<2 mm) using the method of Henderson et al. ( 1972). Each soil sample was subjected to repeated dissolution in 6 M hydrochloric acid at 80 cc followed by 0.5 M sodium hydroxide at 100 cc. After this procedure the sample was dissolved in 30% hydroflurosilicic acid at 15-16 cc. Henderson et al. (1972), went on to further concentrate the silica minerals by heavy liquid separation but this step was not necessary here as sufficient quartz and cristobalite were present for identification using X-Ray Diffraction (XRD). The quantities of quartz and cristobalite in the concentrated sample were estimated with XRD from the 33.4 nm and 40.6 nm reflections respectively. Standards were prepared for both cristobalite and quartz with a matrix of ash from the sampling site which had a negligible content of quartz or cristobalite. The results of the determinations calculated for the bulk soil (<2 mm) are shown in Fig. 4.3A and B. The reflection counts from the standard mixtures were reproducible within 10% or better, giving rise to precision of 10% in the sample determinations. The quantities of plant opal were not estimated ± but its presence was noted in all samples under a polarising microscope. Halloysite was identified in the bulk soil (<2 mm) using the XRD methods described by Churchman et al. (1984). Initially a 101 nm reflection and occasionally a

86 poorly defined peak in the region of 7 4 nm were obtained for ground samples mixed into a slurry and dried on a ceramic tile. After spraying the sample with formamide the 101-102 nm peak sharpened and enlarged considerably and any reflection around 7 4 nm was reduced to background levels. This test indicates that the intermittent 7 4 nm reflection was not due to kaolinite but to a hydrated form of halloysite which upon addition of formamide expanded to give reflections around 102 nm. Following heating of the sample at 110 oc the 102 nm reflection disappeared, and a small peak at 74 nm was seen. The disappearance of the 102 nm peak indicated that mica minerals are not present. To confirm the identification of halloysite the 1-2 )!m fractions of selected ash samples were separated by settling after dispersion of the sample with 1:1 ammonium hydroxide. These samples were then examined under a transmission electron microscope, where the characteristic tubular and spherical morphologies of halloysite particles enabled their identification, after Kirkman (1981 ). The quantity of halloysite was assessed throughout the ash sequence (Fig. 4.3C) using Differential Thermal Analysis (DTA) following the methods of Whitton and Churchman (1987). The depth of the dehydroxylation endotherm at around 540-550 oc was used to estimate the halloysite content. Standards were prepared using an andesite ash matrix with negligible halloysite content. Standard results were reproducible within 10%, giving rise to a precision of ± 10% in the sample estimates. Amorphous constituents of the fine ash material were estimated using two methods. The first is an indirect method for estimating the short range order and organic constituents (SROCO) in a soil (AIIoway et al. , 1992a). Air-dried soil (<2 mm) is shaken in the dark for 24 hours in 0.2 M acid-oxalate reagent (pH 3.0), at a ratio of 100 ml of reagent to 1 g of soil. SROCO content is the weight of material dissolved from the soil. Air dry analysis does not however take into account the weight of water present within the structure of, or held tightly by the short range order minerals such as allophane. A moisture correction factor is then applied here to account for the water held by the air dry soil before and after extraction. The SROCO content of the ash materials (Fig. 4.30) ranges from around 3-9 % of the oven dry bulk soil. The other, more quantitative method used is that of Parfitt and Wilson ( 1985) using oxalate and pyrophosphate extractable chemistry. Bulk soil samples (<2 mm) were extracted with acid-oxalate and pyrophosphate reagents following the procedures of Blakemore et al. (1987). The Si, AI, and Fe content of the acid-oxalate extracts (Sio, Alo, and Feo), and the AI content of the pyrophosphate extract (Alp) were assessed using flame atomic absorption. The acid-oxalate dissolves the short-range order and organic constituents, while the Alp provides an estimate of the aluminium in Al-humus complexes. The established methodology is then to asses the AI/Si ratio for allophane from (Aio-Aip)/Sio multiplied by 28/27 to give the atomic ratio. The Sio content is then multiplied by a factor corresponding to the atomic ratio to give the percentage of allophane in the soil sample (Fig. 4.3E). Ferrihydrite content of the ash materials was estimated (Fig. 4.3F) by multiplying Feo by 1.7 after Childs (1985).

87 The grain size distribution through the fine ash material (Fig. 4.2) was estimated using the method of Alloway et al. (1992a). Short range order and organic constituents (SROCO) are extracted from a sample with 0.2 M acid-oxalate reagent, and crystalline clay, silt and sand fractions are then assessed using a combination of wet and dry sieving. The water content at a matric potential of -1 .5 MPa was determined gravimetrically after 24 hrs soaking and 7 days to equilibrate under the matric potential.

4.9 Results and discussion

The grain size is variable throughout the thickness of ash but shows no trends with depth, apart from the relative abundance of SROCO + crystalline clay near the top of the ash column (Fig. 4.2). The grain size variability reflects additions of air fall ash of varying grades which makes up the sequence. The field moist and air dry gravimetric water content measured at a matric potential of -1 .5 MPa is high to very high for the ash (Table 4.2). The -1.5 MPa water content is highest where the clay + SROCO content is highest.

4.9.1 Primary ash mineralogy

All of the ash samples have been studied under a polarising microscope. Major phases present are plagioclase, orthopyroxene, and clinopyroxene, in order of abundance. The three major phases make up usually >90% by volume of the mineral phases observed. Hornblende and titanomagnetite occur in about half of the samples in quantities usually <5% by volume of the mineral component, but sometimes up to 10%. Highly vesicular and microlite-rich andesitic glass is seen within all samples and comprises over 50% of grains counted in most samples. The andesitic glass is always partially weathered and brown in colour, but not completely replaced by secondary minerals as observed by Kirkman and McHardy (1980) in andesitic glass of tephras from Egmont volcano in western North Island.

4.9.2 Siliceous phases

The quartz content of these ash units is very low throughout the entire range of samples (Fig. 4.3A). The peak of 0.5% quartz at the top of the sampling sequence is due to the presence of the quartz-bearing Okaia Tephra (sourced from Taupo Volcano). No quartz was found in andesitic lapilli units and the small amount of quartz within the ash deposits is here regarded to be a result of aerosolic addition. A small increase in the quartz content occurs near the top of the sequence in a more strongly weathered zone below the Okaia Tephra, with fewer pumice lapilli interbeds present. This suggests a more slowly accumulating soil surface at this time, providing an opportunity for greater weathering and larger quantities of quartz to be deposited onto the soil surface.

88 %Sand %Silt % SROCO Clay + 20 40 60 80 20 40 0 20 40

Figure 4.2. Grain size distribution within the fine ash materials assessed using the method of Alloway et al. (1992a). Coarse ash/lapilli beds and diamictons were not sampled. Ok = Okaia Te phra, Om = Omataroa Tephra, Hu = Hauparu Tephra and Re = Rotoehu Ash. Lithology symbols given in Fig 4.1 .

89 A previous study of aerosolic quartz within andic materials in Taranaki (AIIoway et al., 1992b ), showed two peaks of quartz accumulation corresponding with oxygen isotope stages 2 and 4, and very little quartz in deposits accumulating during o180 stage 3 (c. 25- 60 ka B.P.). Quartz accumulation was correlated with episodes of regional loess deposition throughout the southern North Island in o180 stages 2, and 4. A loess deposition episode during o180 stage 3 did not correspond to a quartz peak in the sequence studied by Alloway et al. (1992b) in Taranaki. The ash deposits studied here on the Ruapehu ring plain accumulated mostly during o180 stage 3 and part of o180 stage 4. Regional loess deposition occurred during both o180 stages 3 and 4, throughout the southern North Island (Milne and Smalley, 1979; Pal mer, 1985; Alloway et al. , 1988; Pillans, 1994 ). Neither of these regional loess deposition episodes are indicated by a quartz peak in the Ruapehu ash sequence even though source areas in the Wanganui hill country to the south-west, or the Kaimanawa Mountains to the east potentially could have supplied quartz to the site. This indicates that the accumulation rate of the ash in this sequence was too high to preserve appreciable quartz and/or the climate was insufficiently severe to mobilise the quartz source. The quartz content in the Ruapehu ring plain ash sequence appears to be a function of the accretion rate of the soil surface rather than of climate change and consequent loess deposition as indicated by the study of Alloway et al. (1992b ). The cristobalite content throughout the ash column (Fig. 4.38) is highly variable from <0.5 to 4%. The origin of cristobalite in volcanic soils has been attributed to both pedogenesis (Lowe, 1986), and as a primary phase from the volcanic parent material (Mizota, et al., 1987). The amount of cristobalite within selected andesite pumice lapilli layers interbedded in the sequence has been estimated using the same methods used for the ash samples. The content of cristobalite in the pumice lapilli was of a similar range to that in the ash material. The cristobalite in the ash material was also in a highly crystalline form as indicated by its sharp X-ray reflections and its presence was noted as microlites within weathered andesitic glass. The cristobalite content of the samples is thus most likely a reflection of the primary composition of the ash, and indicates that the cristobalite content of the erupted ash changed rapidly over time.

4.9.3 Secondarymi nerals

The secondary minerals which have been identified within the ash sequence are halloysite, allophane and ferrihydrite. The halloysite content within the ash sequence ranges from 10 to 25 % of whole soil (<2 mm). The highest contents are found in the lower zone of strong weathering. Halloysite within the deposits occurs in two morphological forms, as hollow tubes and as spheres or ellipsoids. The spherical form is dominant. Kirkman (1981) proposed that the tubular particles formed from feldspar and the spheres from rhyolite glass. Due to the

90 virtual absence of rhyolite glass in this sequence it is considered that the spherical halloysite is here formed from andesitic glass. The allophane content of the ash calculated using the method of Parfitt and Wilson (1985), is very low throughout the sequence (Fig. 4.3E), ranging from 1 to 4%. The trend of allophane content in the sequence is very similar to the SROCO trend, with higher values recorded near the top within the upper zone of strong weathering. The range of estimated ferrihydrite contents is from 0.3 to 1.6%, showing a similar trend to the allophane content (Fig. 4.3F). The calculated allophane and ferrihydrite contents are lower than the SROCO contents of the ash material. Alp content within the ash sequence is barely detectable ranging from 0 to 0.02%, indicating that there is little organic material present which could be making up the rest of the SROCO estimate. This difference was also noted by Alloway et al. (1992b), although they did not take into account the water held by allophane, ferrihydrite and humus complexes in air dried samples. In air dried allophane rich soils significant water is held within the particles and within the structure of SROCO constituents (Table 4.2). This water is measured as a weight difference in the SROCO estimation as the SROCO minerals have been dissolved, but is not seen in the Si and AI content of the extracted solution, leading to overestimation in the SROCO method compared to the Parfitt and Wilson (1985) method. Even after a moisture correction fa ctor is applied to the Ruapehu ash SROCO estimates, they are still higher than the calculated allophane + ferrihydrite. Either the Parfitt and Wilson (1985) empirical method underestimates the allophane within these samples or the acid-oxalate reagent is partially dissolving other phases, contributing to the SROCO estimate. The SROCO extractions were carried out in the dark and past studies have shown that under these conditions very little dissolution of crystalline minerals is expected (e.g. Fey and Le Roux, 1977; Wada, 1977; Parfitt and Childs, 1988). The diffe rences between the two methods is probably due to a combination of, longer extraction times used in the SROCO method (2 successive 24 hour extractions compared to a single 2 or 4 hour extraction), inaccuracies in recovering and weighing the undissolved sample fraction in the SROCO method, and a possible lack of understanding of which phases are being attacked by the oxalate reagent in long extraction times in these tephric samples. The formation of allophane and halloysite from volcanic ash in New Zealand conditions has been addressed in many studies (e.g. Parfitt et al. , 1983 and 1984; Lowe, 1986; Singleton et al. , 1989; Parfitt and Kimble, 1989). The overall conclusion from these studies is that volcanic glass can weather directly to either allophane or halloysite; which mineral is formed depends on the AI/Si ratio of the soil solution. When the AI/Si ratio is low in soil solution, halloysite is formed and when the ratio is high allophane is formed. This model does not require allophane to be formed as an intermediary which later transforms to halloysite, as in the model of Kirkman (1975).

91 %Quartz % Cristobalite % Halloysite % SROCO % Allophane % Ferrihydrite 0 0.3 0.6 0 2.5 5 5 10 15 20 25 0 5 10 0 2 4 0 1 2 (m) A B c D E F

c.o N

35 - lilllf-��::��::��::::��::::::��::��::::::��Figure 4.3. Selected mineralogy of the fine ash sequence. SROCO values have moisture correction applied��'. Ok = Okaia Tephra, Om = Omataroa Te phra, Hu = Hauparu Te phra and Re = Rotoehu Ash. Lithology symbols given in Fig. 4.2. There are many factors which can decrease the AI/Si ratio in soil solution (Lowe, 1986) and thus promote the formation of halloysite from volcanic glass. These factors include low rainfall, poor or impeded drainage, and deep burial of ash deposits. Strong evidence for poor drainage is not seen in the form of common mottling and gleying features in the ash deposits but the dominant brown colouring of the ash deposits could mask all but very strong gleying processes. Some of the diamictons within the sequence have a sandy matrix and may not provide a large impediment to drainage, however others have a clay-rich matrix or are cemented and may impede drainage. The deposits are presently buried by at least 20 m of tephra and diamictons which is likely to have played an important role in their weathering. Dissolved silica has probably leached down into the sequence of ash deposits from the sediments and tephras above. This supply of silica would have promoted the formation of halloysite from the weathering andesitic glass by reducing the AI/Si ratio in soil solution, especially where drainage is impeded. In this sequence it is likely that the effects of both deep and rapid burial of the ash deposits, and poor drainage conditions created by water perching on or below interbedded diamictons have had the greatest effect on formation of secondary minerals. The AI/Si atomic ratio of the allophane present in the ash sequence is 2: 1 or greater throughout the entire sequence. The 2:1 allophane is coexisting in the ash with halloysite. This is at odds with the findings of Parfitt et al., (1984) who found that when allophane and halloysite were coexisting in a rhyolitic ash bed in a moderate or low leaching environment (<200 mm of percolation/year), the allophane had an AI/Si ratio of close to 1:1. Allophane with an AI/Si ratio of 2:1 was formed under conditions of greater leaching (>250 mm a year), and no halloysite was formed under these conditions. The paleosol sequence is not near the current soil surface and modern plant roots do not affect it; thus the allophane is not considered to be modified by modern processes. The allophane may also have lost Si post-formation, thus increasing its AI/Si ratio, but past studies have shown that 2:1 allophane can persist in soils for hundreds of thousands of years in buried soil environments similar to the Ruapehu ring plain sequence (e.g. Kirkman, 1981; Stevens and Vucetich, 1985; Lowe, 1986; Lowe and Percival, 1993), so this affect is not thought to be dominant in the sequence. The coexistence of allophane with a 2: 1 AI/Si ratio and appreciable halloysite in the Upper Waikato Stream ash deposits may indicate that the two minerals were formed at different times in the weathering history of the ash materials due to a change in the weathering environment. If halloysite is forming from the weathering ash while tephra and ring-plain sediments are being deposited above, the 2:1 allophane may be remnant from weathering of the ash material when it was originally in an acid leaching environment, at or near the soil surface. If this is the case then the trends shown in the allophane content may indicate the length of time the soil surface was stable, or a record of the accumulation rate. Where the soil surface is stable for a longer time more allophane is formed. This agrees with field observation of stronger weathering features and a low frequency of pumice lapilli additions where there is greatest allophane content in the top of the ash sequence. This also correlates with the preferential concentration of quartz in

93 the sequence at this point, due to slower accumulation rates and more intensive weathering. The trend of ferrihydrite contents of the ash parallels that of allophane, and indicates that the ferrihydrite is likely to have been formed when the ash was at or near the soil surface at the same time as allophane. The lower zone of observed relatively strong weathering within the fine ash does not show a peak in allophane or ferrihydrite content. This could be due to the soil surface accumulating too quickly to produce favourable conditions for the formation of allophane and ferrihydrite. Field observation at this point shows that there is a much greater frequency of pumice lapilli additions to the sequence at this time, indicating a relatively high rate of deposition. In this sequence of andesitic tephra with very little rhyolitic glass present, the AI/Si ratio of the soil solution appears to be driving the formation of the secondary minerals as fo und by Singleton et al. (1989), rather than the AI/Si ratio of the parent glass as found by Kirkman and McHardy (1980). The parent glass composition obviously has an effect on the soil solution composition. As it weathers, andesite glass will release more AI into solution than rhyolite glass reflecting their initial AI content diffe rences. The result of this is observed in soils from Taranaki which contain large quantities of allophane (AIIoway et al., 1992b ). In the ash materials on the Ruapehu ring-plain sequence the effect of the weathering andesitic glass composition on the soil solution is considered to be overshadowed by the other environmental factors of rapid burial and impeded drainage. This is reflected by the dominance of halloysite as a weathering product and only a limited time for allophane formation when the ash material was at or near the surface.

4.10 Relationship of Ruapehu ring-plain sequence to climate record for southern North Island

The observed features of stronger weathering and more abundant root rhizomorphs within parts of the ash sequence could be indicating the climate was warmer during these periods of deposition than at other periods within the sequence. The effect of climate change during the period of deposition on these ash deposits would have been overprinted onto the effect of accumulation rate discussed previously. The timing and nature of climate oscillations within southern North Island, New Zealand has been well documented (Palmer, 1985; Alloway et al., 1992b and 1992c; Pillans, 1994), and landscape events can be shown to be related to the marine o180 record (Fig. 4.4). The presence of identified rhyolitic tephras provide time planes which enable correlation of features in the ash sequence to the established climate record of the southern North Island. The range in age of the deposits within this sequence is from c. 24 to 70 ka, mostly during o180 stage 3. Within the upper zone of strong weathering, the presence of the Omataroa Tephra (ea. 28.2 ka; Table 4.1) correlates this weathering period with a widespread soil development episode in loess sequences of the southern North Island (Fig. 4.4; Milne and Smalley, 1979 ; Leamy et al., 1973; Palmer, 1985; Alloway et al., 1988; Pillans 1988).

94 Age M.O.I. stages Rangitikei Taranaki Wanganui NE Ruapehu (ka)

Sy1 L1 2 20 ffi:{J Diamictons Whanahuia r Paleosol 30 11;r; � 11 IIIAt;ll Andesitic D tephra L-----1-- Hu � Aggradation 40 Sy2 L2 � gravels

(0 CJ'1 3 Kimbolton Paleosol Loess 50 D

111111Sr3 IllS3I Strongest paleosol - development 60 111111111 11111111111 11111111111 Ill11 1111 11 Re

70 4 Sy3 L3

Figure 4.4. The relationship of the northeastern Ruapehu ring-plain sequence with deep sea 180 isotope record (M.O.I. Stages; Shackleton et a/., 1990), and southern North Island loess stratigraphy from the Rangitikei valley (leamy et al., 1973; Milne and Smalley, 1979), Wanganui (Pillans, 1988; 1994), and Taranaki (AIIoway et al., 1988). Oh = Ohakea loess, Rt = Rata loess and Po = Porewa loess L 1 = loess 1, S2 = buried soil 2 etc. Sy1 = loess 1 correlative, Sr2 = buried soil 2 correaltive etc. Kk = Kawakawa Tephra, Ok = Okaia Te phra, Om = Omataroa Te phra, Hu = Hauparu Tephra and Re = Rotoehu Ash. This soil development is represented by one of the weakest of the Last Glacial paleosols. In the ash sequence from eastern Ruapehu the weathering of this paleosol is further enhanced by relatively slow accretion of the soil surface as discussed previously. Within the lower zone of strong weathering, the presence of the Rotoehu Ash (ea. 64 ka; Table 4.1) indicates this weathering episode was contemporaneous with another period of widespread soil development on loess throughout the southern North Island (Fig. 4.4; Milne and Smalley, 1979; Leamy et al. , 1973; Palmer, 1985; Alloway et al., 1988; Pillans, 1988). This soil development is represented by very strong paleosols throughout the southern North Island. In the eastern Ruapehu sequence the intensity of soil development was probably hindered by rapid accretion of tephra at the soil surface. The cool climatic episode in late 8180 stage 3, represented by loess deposition in the southern North Island (Fig. 4.4), is not evidenced in the Ruapehu sequence. Quartz input is swamped by a high accretion rate of andesite ash and lapilli. The only manifestations of a cool climate are less observable weathering features, less root rhizomorphs, and lower amounts of allophane and ferrihydrite in the ash deposited during this time. There were intervals of slow deposition within this time frame evidenced by low frequencies of pumice lapilli layers at various levels in the sequence. In these parts of the sequence the weathering development may have been slow under a cool and possibly dry climate.

4.11 Conclusions

The provenance of ash material within the sequence is almost entirely local, and represents continuous deposition of andesite ash and lapilli. If there is any redeposited material of ash grade within the sequence it is very locally derived volcanic loess blown from other parts of the ring plain. The input of material from outside of the local area such as regional loess deposition recorded in other parts of the southern North Island is very low. Although source areas proximal to the study area exist {< 10 km away), aerosolic quartz input into this sequence appears to have been diluted by an overwhelming component of rapidly accumulating locally derived ash and lapilli. Weathering zones within the ash materials on the eastern Ruapehu ring plain are correlated to mild climate episodes during the Last Glacial, contemporaneous with periods of widespread soil development on loess throughout the southern North Island (Fig. 4.4; Leamy et a/, 1973; Palmer, 1985; Alloway, et al., 1988; Pillans, 1988). However, weathering features within the ash materials on the eastern Ruapehu ring plain are not only a function of the climate when the ash was accumulating, but are also strongly dependent on the accretion rate of the soil surface. Slow tephra accretion enhanced weathering features during the development of the soil on fine ash deposited during late 8180 stage 3. Rapid tephra accretion during the development of the soil in fine ash deposited during early 8180 stage 3 may have reduced the expression of weathering features and secondary minerals in the ash at this time.

96 The weathering minerals within this ash sequence appear to have been formed in two separate periods in the history of the ash sequence. This is deduced from the coexistence of halloysite in appreciable quantities (1 0-25% of bulk soil <2 mm) with allophane which has an AI/Si atomic ratio of >2: 1. Allophane is present in small quantities according to estimates using the Parfitt and Wilson (1985) method. The sequence can be regarded as an aggrading sequence of Entisols and Andisols in which the rapid accretion of the soil surface is hindering further development of the soil. The allophane and ferrihydrite content of the ash material shows a distribution which has an inverse relationship with the accumulation rate indicated by quartz content, field observations of pumice lapilli frequency and weathering features. The allophane and ferrihydrite content is greatest where the accumulation rate of the ash appears lowest. This indicates that allophane and ferrihydrite in this sequence were formed from weathering andesitic glass when the ash material was at or near the surface. Subsequent, rapid burial of the ash and poor drainage conditions in the sequence probably hindered further formation of allophane as silica leached from the deposits above to decrease the AI/Si ratio in soil solution and thus prevent further allophane formation (Parfitt et al., 1984; Singleton et al., 1989). Halloysite probably formed after the burial of the ash materials. As silica was leached down from tephra and diamictons above, the AI/Si ratio in the soil solution was decreased, and the fo rmation of halloysite was promoted from weathering andesitic glass (Parfitt et al., 1984; Singleton et al. , 1989). The peaks in concentration of halloysite within the ash profile may relate to where the drainage is impeded as water and silica move down the profile. The drainage impediment could be a textural change within the ash or a diamicton interbedded within the sequence. The highest contents of halloysite occur beneath a large cemented diamicton unit (Fig. 4.1 }, exemplifying this interpretation. The now buried ash sequence represents a continuously accreting Entisoi/Andisol. The weathering features and secondary minerals within the sequence are a function of a complex mix of accretion rate, late Quaternary climate change, and post-depositional weathering.

4.12 Acknowledgements

This work forms part of the Ph.D. research of Shane Cronin and we gratefully acknowledge funding from the New Zealand Vice Chancellors Committee, Massey University Graduate Research Fund, the Helen E. Akers Scholarship fund, and the Department of Soil Science of Massey University. J.S. Whitton is thanked for his help with mineral analyses, J.H. Kirkman and R.B. Stewart are thanked for their comments on an earlier version of the manuscript. We also wish to thank B.V. Alloway and R.L. Parfitt for their comprehensive reviews of the manuscript.

97 4.13 References

Alloway, B.V., Neall, V.E. and Vucetich, C.G., 1988. Localised loess deposits in north Taranaki, North Island, New Zealand. In: Loess - Its distribution geology and soils, Eden, D.N., and Furkert, R.J. (eds). A.A. Balkema Publishers, Rotterdam. Alloway, B.V., Neall, V.E. and Vucetich, C.G., 1992a. Particle size analyses of Late Quaternary allophane-dominated andesitic deposits from New Zealand. Quat. lnt., 13/14, 167-1 74. Alloway, B.V., Stewart, R.B., Neall, V.E. and Vucetich, C.G., 1992b. Climate of the Last Glaciation in New Zealand, based on aerosolic quartz influx in an andesitic terrain. Quat. Res., 38, 170-179. Alloway, B.V., McGione, M.S., Neall, V.E. and Vucetich, e.G., 1992c. The role of Egmont-sourced tephra in evaluating the paleoclimatic correspondence between the bio- and soil-stratigraphic records of central Taranaki, New Zealand. Quat. lnt., 13/14, 187-194. Blakemore, L.C., Searle, P.L. and Daly, B.K., 1987. Methods for chemical analysis of soils. N. Z. Soil Bureau Scientific Rep., 80. Childs, C.W., 1985. Towards understanding soil mineralogy. 11 Notes on ferrihydrite. N. Z. Soil Bureau Lab.Rep., CM?, DSIR, New Zealand. Churchman, G.J., Whitton, J.S., Claridge, G.G.C. and Theng, B.K.G., 1984. Intercalation method using formamide for diffe rentiating halloysite from kaolinite. Clays and Clay Min., 32, 241 -248. Cronin, S.J., Neall, V.E. and Palmer, A.S., 1996. The geological history of the north eastern ring plain of Ruapehu Volcano, New Zealand. Quat. lnt., 34-36: 21-28. Fey, M, and Le Roux, J., 1977. Properties and quantitative estimation of poorly crystalline components in sesquoxidic soil clays. Clays and Clay Min., 25, 285- 294. Flint, R.V., 1960. Diamictite, a substitute term for symmictite. Geol. Soc. of America Bull., 71, 1809. Froggatt, P.C. and Lowe, D.J., 1990. A review of late Quaternary silicic and some other tephra formations from New Zealand: their stratigraphy, nomenclature, distribution, volume, and age. N. Z. J. of Geol. and Geophys., 33, 89-109. Hackett, W.R. and Houghton, B.F., 1989. A facies model for a Quaternary andesitic composite volcano: Ruapehu, New Zealand. Bull. Volcano!., 51, 51-68. Henderson, J.H., Clayton, R.N., Jackson, M.L., Syers, J.K., Rex, R.W., Brown, J.L., and Sachs, I.B., 1972. Cristobalite and quartz isolation from soils and sediments by hydroflurosilicic acid treatment and heavy liquid separation. Soil Sci. Soc. America Proc., 36, No. 5, 830-835. Kirkman, J.H., 1975. Clay mineralogy of some tephra beds of the Rotorua area, North Island, New Zealand. Clay Min., 10, 437-449. Kirkman, J.H., 1981 . Morphology and structure of halloysite in New Zealand tephras. Clays and Clay Min., 29, 1-9.

98 Kirkman, J.H., and McHardy, W.J., 1980. A comparative study of the morphology, chemical composition and weathering rhyolite and andesite glass. Clay Min., 15, 165-173. Leamy, M.L., Milne, J.D.G., Pullar, W.A. and Bruce, J.G., 1973. Paleopedology and soil stratigraphy in the New Zealand Quaternary succession. N. Z. J. of Geol. and Geophys., 16, 723-744. Lowe, D.J., 1986. Controls on the rates of weathering and clay mineral genesis in airfall tephras: a review and New Zealand case study. In: Colman, S.M., and Dethier, D.P. (eds) Rates of chemical weathering of rocks and minerals. Academic Press, Orlando, pp. 265-330. Lowe, D.J. and Percival, H.J., 1993. Clay mineralogy of tephras and associated paleosols and soils, and hydrothermal deposits, North Island. Guide book for New Zealand pre-conference field trip F.1. 1Oth International Clay Conference, Adelaide, Australia. McGione, M.S. and Topping, W.W., 1977. Aranuian (post-glacial} pollen diagrams from the Tongariro region, New Zealand. N. Z. J. of Botany, 15, 749-760. Milne, J.D.G. and Smalley, I.J., 1979. Loess deposits in the southern part of the North Island of New Zealand: an outline stratigraphy. Acta Geologica Academiae Scientarium Hungaricae, 22, 197-204. Mizota, C., Toh, N. and Matsuhisa, Y., 1987. Origin of cristobalite in soils derived from volcanic ash in temperate and Tropical regions. Geoderma, 39, 323-330. New Zealand Meteorological Service, 1973. Rainfall normals for New Zealand, 1941- 1970. Mise. Publ., 145. Palmer, A.S., 1985. The Last Interglacial record in Wairarapa Valley, New Zealand. In: Pillans, B.J. (ed). Proceedings of the second CLIMANZ conference 1985. Publication No.31 of Geology Department, Victoria University of Wellington, N.Z. Parfitt, R.L. and Childs, C.W., 1988. Estimation of forms of Fe and AI: A review, and analysis of contrasting soils by dissolution and Moessbauer methods. Australian J. of Soil Res., 26, 121 -144. Parfitt, R.L. and Kimble, J.M., 1989. Conditions for formation of allophane in soils. Soil Sci. Soc. of America J., 53, 971-977. Parfitt, R.L., Russell, M. and Orbell, G.E., 1983. Weathering sequence of soils from volcanic ash involving allophane and halloysite, New Zealand. Geoderma, 29, 41- 57. Parfitt, R.L., Saigusa, M. and Cowie, J.D., 1984. Allophane and halloysite fo rmation in a volcanic ash bed under different moisture conditions. Soil Sci., 138, 360-364. Parfitt, R.L., and Wilson, A.D., 1985. Estimation of allophane and halloysite in three sequences of volcanic soils, New Zealand. Catena Suppl. 7, 1-8. Pillans, B.J., 1988. Loess chronology in Wanganui Basin, New Zealand. In: Loess - Its distribution geology and soils, Eden, D.N., and Furkert, R.J. (eds). A.A. Balkema Publishers, Rotterdam.

99 Pillans, B.J., 1994. Direct marine-terrestrial correlations, Wanganui Basin, New Zealand: the last 1 million years. Quat. Sci. Reviews, 13, 189-200. Shackleton, N.J., Berger, A. and Peltier, W.R., 1990. An alternative astronomical calibration of the lower Pleistocene timescale based on ODP Site 677. Trans. Ray. Soc. of Edinburgh: Earth Sci., 81, 251-261. Singleton, P.L., Mcleod, M. and Percival, H.J., 1989. Allophane and halloysite content and soil solution silicon in soils from rhyolitic volcanic material, New Zealand. Australian J. of Soil Res., 27, 67-77. Stevens, K.F., and Vucetich, G.C., 1985. Weathering of Upper Quaternary tephras in New Zealand. Part 11 Clay minerals. Chem. Geol., 53, 237-247. Wada, K., 1977. Allophane and lmogolite. In: Minerals in Soil Environments, Dixon, J.B., and Weed, S.B. (eds), 603-638. Soil Sci. Soc. of America, Madison, Wisconsin. Whitton, J.S. and Churchman, G.J., 1987. Standard methods for mineral analysis of soil survey samples for characterisation and classification in New Zealand Soil Bureau. N.Z. Soil Bureau Sci. Rep., 79. Wilson, C.J.N., Switzur, R.V. and Ward, A.P., 1988. A new 14C age for the Oruanui (Wairakei) eruption, New Zealand. Geol. Mag., 125, 297-300. Wilson, C.J.N., Houghton, B.F., Lanphere, M.A. and Weaver, S.D., 1992. A new radiometric age estimate for the Rotoehu ash from Mayor Island volcano, New Zealand. N. Z. J. of Geol. and Geophys., 35, 371-374.

100 CHAPTER 5: THE LAHAR RECORD AND CONSTRUCTION OF THE NORTHEASTERN RUAPHEU AND EASTERN TONGARIRO RING PLAINS

5.1 Introduction

The NE Ruapehu and E Tongariro ring plains are composed of laharic, fluvial and glacial sediments as well as tephra and lava lithologies. However, the ring plains are volumetrically dominated by volcaniclastic diamictons deposited by lahars. Thus by dating and mapping these lahars, the timing and landform construction of much of the ring plain areas can be established. In addition, the timing of lahars can be used to assess the lahar hazard, which is contained in the following chapter. The rhyolitic and andesitic tephra stratigraphy of Topping (1973), Donoghue et al., (1995), and that established in Chapters 2 and 3 were used to date and map lahar deposits on the ring plain areas studied. Once the lahar deposits were dated and mapped, the distribution and timing of lahars in conjunction with information from the paleosol record developed in Chapter 4 was used to elucidate the construction of the ring plain areas studied. In this way a history of the timing of construction of various ring plain areas was established. In addition, the influence of climatic and volcanic events on the ring plain construction is explained. This part of the study is in press within the New Zealand Journal of Geology and Geophysics, Vol. 40, No. 2 (1997). The contributions of the two authors to the study were as follows:

S.J. Cronin: Principal investigator Carried out all: field descriptions and sampling tephra and lahar correlation and mapping preparation and writing of the manuscript

V.E. Neall: Adviser Aided the study by: discussion of results and methodology editing and discussion of the manuscript

101 5.2 A LATE QUATERNARY STRATIGRAPHIC FRAMEWORK FOR THE NORTHEASTERN RUAPEHU AND EASTERN TONGARIRO RING PLAINS, NEW ZEALAND

Shane J. Cronin and Vincent E. Neall Department of Soil Science, Massey University, Private Bag 11 222, Palmerston North, New Zealand

New Zealand Journalof Geology and Geophysics 40, No. 2.

Abstract The northeastern Ruapehu and eastern Tongariro ring plains record a complex sequence of episodic lahar sedimentation. Using andesitic and rhyolitic tephrostratigraphy, fifteen lahar episodes are recognised in the northeastern Ruapehu ring-plain record ranging in age from >65 to 5 ka, and five in the Tongariro ring-plain record ranging in age from >23 to 14 ka. The most voluminous and widespread lahar deposition occurred during cool and stormy climatic periods equivalent in age to marine 8180 stages 2 and 4. In these periods, accelerated physical weathering on the volcanoes supplied erosion debris, while large areas of snow and ice acted as sources of water to form lahars triggered by a variety of mechanisms. Lahar distribution after c. 22.5 ka was affected by two landform changes in the area at about this time. First, a large lava flow was emplaced along the boundary between Ruapehu and Tongariro ring plains shortly before 22.5 ka, effectively separating the two ring plains since. Second, Last Glacial moraines along the Whangaehu and Mangatoetoenui Rivers have blocked direct drainage from the Ruapehu summit region to a large sector of the northeastern ring plain, including the Upper Waikato Stream, formerly an important lahar path. These moraines have directed subsequent lahars (after c. 15 ka) along their current routes (active in 1995) in the Whangaehu and Mangatoetoenui catchments. Lahar deposition in the study area during this time is linked to large, explosive andesitic eruptions impacting on catchments where retreating glaciers provided water for lahar generation.

5.3 Keywords Ruapehu volcano, ring plain, volcanic stratigraphy, lahars, Late Quaternary, paleoclimatic change.

5.4 Introduction

Ring-plain sediments of composite volcanoes provide a valuable record of past eruptive and sedimentary events that may not be preserved closer to source. The sedimentary record of the ring plain can often be dated precisely using interbedded tephras and radiometric techniques to provide a stratigraphic record of such events missing on the volcano. In the central North Island of New Zealand, coalescing ring plains of the eastern flanks of Tongariro and Ruapehu volcanoes are incised by numerous stream channels

102 that expose a detailed record of their ring-plain sediments and structure. A well established rhyolitic tephra record from the North Island (Froggatt and Lowe, 1990) and the developing andesitic tephrochronology in the Tongariro Volcanic Centre (TgVC - Topping, 1973; Donoghue et al. , 1995; Cronin et al., 1996(a)) provide a good chronological framework for the ring-plain deposits. The aim of this study was to elucidate the stratigraphy and construction of the northeastern Ruapehu and eastern Tongariro ring plains, based on detailed investigations of sections in streams draining the eastern flanks of the volcanoes.

5.5 Setting

Tongariro and Ruapehu are the two largest and most recently active volcanoes of TgVC (Gregg, 1960). Ruapehu is presently active. lt is, the highest peak in the North Island at 2797 m, comprising a 110 km3 composite cone with a ring plain of similar volume surrounding it (Hackett and Houghton, 1989). Tongariro volcano is a slightly smaller massif constructed from several coalescing volcanic cones, the largest of which is the recently active cone of Ngauruhoe (Mathews, 1967). Tongariro volcano is also surrounded by an extensive ring plain. The ring-plain deposits of these volcanoes are confined in the east by the Kaimanawa Mountains and extend down drainage channels to the north and south. Sections examined in this study are along streams indicated in Fig. 5.1.

5.6 Time control

Eighteen dated rhyolitic tephras erupted from Taupe Volcanic Zone calderas to the north are preserved within the TgVC ring-plain sediments {Topping and Kohn, 1973; Donoghue et al., 1995; Cronin et al. , 1996(a)). The rhyolitic tephras are identified using combinations of their stratigraphic position, field appearance, mineralogy, and major element glass chemistry. In addition, dated andesitic tephras erupted from Tongariro and Ruapehu volcanoes provide valuable stratigraphic markers {Topping, 1973; Donoghue, et al., 1995). They are identified using their characteristic field appearances, stratigraphic position and occasionally their mineralogy. The names and ages of marker tephras used in this study are listed in Table 5.1.

5.7 Terminology and identification of ring plain deposits

Ring-plain sediments range greatly in their nature and mode of deposition. At one end of the depositional series are well-bedded silts, sands, and gravels deposited by streamflow mechanisms. At the other end of the series are poorly sorted, matrix-supported units deposited by debris flow mechanisms. A whole range of deposits intermediate in character occur between these two extremes.

103 175° 45'

\ / \ Group 3 // I I

North I Island \ I I Mt To ngariro .A. I I M! Ng )ho � � I 1500 m 39° 10 I

Group 2. ... · ,. I

\ I

I 1 Group 1

I I Mt Ruapehu I ....

0 5km - State Highways Figure 5.1 Location of To ngariro Volcanic Centre including Mts. To ngariro, Ngauruhoe and Ruapehu, and the streams dissecting the eastern Ruapehu and To ngariro ring plains. The streams are discussed in the text in three groups; labelled Groups 1, 2 and 3 on the map.

104 Table 5.1. Tephra coverbeds used in this study.

Te�h ra la:ters Abbreviation Source• Age {ka} Reference to age

Ngauruhoe Tephra Ng TgV ea. 1.8-01 Donoghue (1991)

Taupo Tephra (Unit Y) Tp TVC 1.852 Froggatt and Lowe (1990)

Mangatawai Tephra Mgt TgV 2.5-1.852 Donoghue (1991)

Waimahia Tephra (Unit S) Wm TVC 3.32 Froggatt and Lowe (1990)

Hinemaiaia Tephra (Units 1-R) Hm TVC 5.2-3.952 Wilson (1993)

Motutere Tephra (Units G-H) Mt TVC 5.8-5.32

Mangamate Tephra (Mm), Poutu Pt TgV ea. 9.J1 Topping (1973) Lapilli Mm, Wharepu Tephra Wp TgV ea. 9.71

Poronui Tephra (Unit C) Po TVC 9.812 Froggatt and Lowe (1990)

Mm, Ohinepango Tephra Oh TgV ea. 9.71 Topping (1973)

Mm, Waihohonu Lapilli Wa TgV ea. 9.71

Mm, Oturere Lapilli Ot TgV 9.782

Karapiti Tephra (Unit B) Kp TVC 10Y Wilson (1993)

Pahoka Tephra Pk TgV ea. 10-9.81 Topping (1973)

Bullot Formation (Bt), Pourahu Ph RV ea. 101 Donoghue (1991) Member Waiohau Tephra Wh 11.852 Froggatt and Lowe (1990) ovc Rotorua Tephra Rr 13.082 ovc Rotoaira Tephra Ra TgV 13.82 Topping (1973)

Rerewhakaaitu Tephra Rk 14.72 Froggatt and Lowe (1990) ovc Kawakawa Tephra Kk TVC 22.592 Wilson et al., (1988)

Okaia Tephra Ok TVC ea. 231 Froggatt and Lowe (1990)

Omataroa Tephra Om 28.222 ovc Hauparu Tephra Hu 35.872 ovc Rotoehu Ash Re 643 Wilson et al., (1992) ovc a TgV = Tongariro volcano, RV = Ruapehu volcano, TVC = Taupo Volcanic Centre, OVC = Okataina Volcanic Centre. 1 Estimated ages. 2 Average or single 14C ages on old (Libby) half life basis.

3 Whole rock K-Ar age.

We use the Smith and Fritz ( 1989) definition of the term "lahar" in this study as: "a rapidly flowing mixture of rock debris and water (other than normal streamflow) from a volcano". This definition encompasses volcano-derived hyperconcentrated streamflows and debris flows. The definition is restricted to the process and does not include the deposit. The use of the terms "debris flow" and "hyperconcentrated" streamflow follow that of Smith

105 (1986), Pierson and Costa (1987), and Scott (1988). Table 5.2 outlines the characteristic features of the sediments within the ring-plain sequence, and relates their observed lithologies and sedimentary features to their interpreted emplacement mechanisms.

Table 5.2. Sedimentary features of ring plain deposits and their inferred mode of deposition.

Sediment type Observed features Inferred mode of deposition (after Smith, 1986)

Bedded sands, silts -Variable grainsize, silt, sand and gravels in Normal streamflow and gravels layers and lenses -Individual layers and lenses well sorted -Strong development of dominantly wavy and cross bedding -Ciast supported -Maximum clast diameter 0.2 m

Horizontally/planar -Grain size dominantly med-coarse sand, Hyperconcentrated bedded sands supporting common pebbles and rare streamflow cobbles -boulders -Poorly sorted -Ciast supported -Horizontal bedding discontinuous but overall planar or horizontal fabric -Larger clasts often in pebble or cobble "strings" -Maximum clast diameter 1 m

Sandy-matrix -Very poorly sorted Debris flow massive -Matrix supported diamictons -Matrix dominantly med-coarse sand -unbedded, mostly ungraded and occasionally normally graded -Maximum clast diameter 1-3 m

Silty-matrix -Features as above except matrix Debris flow massive dominated by silt and clay diamictons -can contain hydrothermally altered pumice and lithic clasts -ungraded, and normally graded

5.8 Present channel geography

A large number of sections were examined in several streams. Measured sections at different locations along the length of each stream (e.g. Mangatoetoenui Stream Fig. 5.2) were compared and correlated to obtain a composite stratigraphic column for each stream (Fig. 5.3, 5.5 and 5.6). The streams can be split into three catchment groups. Group 1 comprises Upper Waikato, Tangatu, Wharepu, Te Piripiri, Mangatoetoeiti and Mangatoetoenui Streams (Fig. 5.1) which drain the northeastern flanks of Ruapehu volcano. Ohinepango and Waihohonu Streams form Group 2 which drain a sector of northeastern Ruapehu volcano as well as the southeastern sector of Tongariro volcano. Group 3 comprises Oturere, Makahikatoa, Mangatawai, Mangamate, Mangahouhounui,

106 and Tauhurangi Streams which drain the eastern and northeastern flanks of Tongariro volcano. The three stream catchment groups receive lahars from different source areas and potentially provide different sets of information about the sources, timing, and distribution of lahar events on the eastern flanks of the two volcanoes.

5.9 Group 1 - Northeastern Ruapehu streams (Fig. 5.3)

A clear boundary exists between drainage from the northeastern and southeastern flanks of Ruapehu. To the southeast a large active fan of Holocene laharic and stream sediments has been formed by the Whangaehu River (Palmer et al., 1993). The fan sediments have been deposited by lahar events that are channelled down the upper part of Whangaehu River and avulse their courses on the relatively unconfined lower ring plain (Palmer et al., 1993). This route has been the major path for Ruapehu lahars in historic times (e.g., 1953, 1969, and 1975; Gregg, 1960; Nairn et a/. , 1979), including 1995. All drainage channels north of the Whangaehu River drain northeast into the Tongariro River. There is no historical evidence of lahars overflowing northward from the Whangaehu fan into the Tongariro. However, restricted lahar deposits younger than c. 2 ka are present in the upper portions of the Upper Waikato Stream, and are thought to originate from the Whangaehu catchment (Donoghue, 1991 ). This suggests that some lahars may have overflowed into the Upper Waikato Stream in the recent geologic past. Of the northeastern Ruapehu streams, the Mangatoetoenui is the only channel to have an historic record of lahars (Nairn et al. , 1979), including at least one lahar in the 1995 eruptive sequence. A correlation diagram for the sequences in the six main drainage channels from northeastern Ruapehu is presented in Fig. 5.3. Fifteen episodes of lahar deposition are recognised and numbered consecutively R01-R15 from youngest to oldest. These episodes are recognised by debris flow and hyperconcentrated streamflow deposits, each episode containing a single lahar or series of lahars within a short time interval. The stratigraphic ages of the diamicton packages are determined from their tephra interbeds (Fig. 5.4). Upper Waikato and Tangatu Streams contain the longest stratigraphic records, although Tangatu Stream sections show evidence of extensive coverbed stripping and erosional unconformities. The surfaces around Tangatu Stream do not appear to have stabilised until immediately before the eruption of the Hinemaiaia Tephra, c. 5 ka B.P. In Upper Waikato Stream, the oldest and also the thickest diamicton sequence recorded is R15, which comprises the deposits of several lahars. Rotoehu Ash (c. 64 ka) occurs as glass shards scattered throughout fine andesitic ash overlying lahar deposits of episode R14, thus providing a minimum age for R14 and R15. R13 comprises the deposit of a single lahar whilst R12 contains deposits of two. Andesitic pumice lapilli are abundant within both R12 and R13 and as a pinching shower-bedded tephra between them. R1 1 consists of a complicated collection of deposits of numerous lahars.

107 km from source km from source km from source km from source km from source km from source km from source 15 16 17.5 18.1 18.7 19.5 20.1 T20/440143 T20/449148 T20/461154 T20/465156 T20/470158 T20/476161 T20/480164

Tp Mt I'" 1 I I I I -- Pt Mm r===� Mm �;{j�

-->.

0 (X)

Pk Key to lithology

t=;:·.p•nr...... Andesitic lava Ros lgnimbrite Pahoka and Pourahu Tephras Andesitic tephras Bedded sands, silts and gravels R06-09 u = Unconformity 2m Sandy matrix debris flow deposits Hyperconcentrated streamflow deposits Silty matrix debris flow deposits I Lignite

Figure 5.2 A Sequence of selected measured sections in the direction of stream flow in the Mangatoetoenui Stream. These sections and others were used to produce the composite section shown in Fig. 5.3. Upper Waikato Ta ngatu Stream Wharepu Stream Te Piripiri Stream Te Piripiri Stream Mangatoetoeiti Mangatoetoenui Stream South Tributary Main Tributary Stream Stream

- �k F4 ,. h l = �� Rk Kk Mt I I Pt R1 0 ...... -u '" Bt l - - . · . Wh - .�. 1!33: R11 cr"J �r \ � Ok Oh Wa · .1 � 1x:z:. = 0 CD

m Key to lithology } Andesitic lava lgnimbrite Pahoka and Pourahu Te phras Andesitic tephras Bedded sands, silts and gravels Sandy matrix debris flow deposits Hyperconcentrated streamflow deposits Silty matrix debris flow deposits Figure Composite stratigraphic sections for Group streams draining the northeastern Lignite 5.3 flanks of Ruapehu. Tephra abbreviations are1 listed in Table 5.1, lahar deposit nomenclature from text. u = Unconformity Other than in the Upper Waikato and Tangatu Streams, lahar deposits of episodes older than R1 0 are only well preserved along the Desert Road Fault scarp beside the Whangaehu River. The youngest lahar deposit in the Upper Waikato Stream sequence is from the R1 0 episode. This is part of an overall aggradation package of sediments deposited during the Last Glacial Maximum. The R10 deposit is a lithologically distinctive unit. lt has a yellowish brown to brown silt and clay-rich matrix which supports pebbles to large boulders. These clasts are subangular to angular, multi-coloured and multi-lithologic, including abundant, red, yellow and white, hydrothermally altered soft clasts. In many places this deposit comprises two or three flow units separated by thin (5 mm) clay laminae. The uppermost flow unit exhibits crude normal grading with large boulders concentrated at the base. The lower one or two flow units are finer grained and ungraded. Unit R1 0 is generally the lowest unit exposed in the streams north of Tangatu Stream. The stream sections between Tangatu and Mangatoetoenui Streams collectively record nine episodes of lahar deposition younger than R1 0. Wharepu Stream and small surrounding streams, including the southern tributary of the Te Piripiri, all record essentially the same sequence, R10 and R09. The R09 sequence comprises the deposits of several lahars below the Rerewhakaaitu Tephra. The main Te Piripiri Stream channel records the southernmost occurrence of six lahar deposits younger than R09. These six units (ROB, R07, R06, R05, R04 and R02) are all deposits of single lahars and are well constrained in age by enclosing tephra layers (Fig. 5.4 ). Units R04 and R05 are correlated to the fragmentary sequence within Managatoetoeiti Stream and R04-R09 can be traced into Mangatoetoenui Stream sections. In the latter sequence, two further lahar deposits are preserved; R03, together with the youngest recorded episode R01 (5.2-5.3 ka).

5.10 Group 2 - Streams draining both volcanoes (Fig. 5.5)

The Ohinepango and Waihohonu Streams drain the area between the Tongariro and Ruapehu massifs and have the potential to receive sediment from both volcanoes. The records in these streams are complicated by the presence of a large lava flow and its associated autobreccia deposits. The composite stream section fo r Ohinepango Stream (Fig. 5.5) shows a sequence of lahar deposits, which can be correlated to units in the neighbouring Mangatoetoenui Stream. Most notably, the distinctive R1 0 debris flow deposit is present below the Kawakawa Formation and the lava flow. Sections showing the stratigraphic relationships between the lava flow and Kawakawa Formation occur in the adjacent Waihohonu Stream. The age of the lava flow is well constrained between R1 0 (and the underlying Okaia Tephra) and the Kawakawa Formation, i.e. between c. 23 ka and 22.6 ka B.P. The Ohinepango Stream sections also record R1 1, RO?, R06, and R04 lahar deposits (Fig. 5.4). The Waihohonu Stream has a much larger proportion of its headwaters on the slopes of Tongariro volcano. Here there are more fluvial deposits

110 throughout the sequence, as well as three lahar deposits which are equivalent in age to R09, RO? and R06 (Fig. 5.4).

Figure 5.4 Stratigraphy of lahar deposition episodes on the north-eastern Ruapehu and eastern Tongariro ring plains.

Marker tephra layers Tephra ages (ka) Group 1 Group 2 Group 3 Northeastern Streams draining EasternTongar iro Rua�ehu streams both volcanoes streams

Hinemaiaia Tephra 5.2-3.95 R1 Motutere Tephra 5.8-5.3 R2 Mm, Poutu Lapilli c. 9.7 Mm, Wharepu Tephra c. 9.7 R3 Mm, Ohinepango Tephra c. 9.9 Mm, Waihohonu Lapilli c. 9.9 R4 R4 Mm, Oturere Lapilli c. 9.9 Karapiti Tephra 10.1 R5 Pahoka Tephra c. 10 R6 R6 Bt, Pourahu Member c. 10 R7 R7 Waiohau Tephra 11.85 R8 R8 T1 Rotorua T ephra 13.08 Rotoaira Tephra 13.8 T2 Rerewhakaaitu Tephra 14.7

R9 R9ff3 + Hinuera T3 + Hinuera Formation Formation Kawakawa Tephra 22.59 R10 R10 T4 R11 R11ff4 T4 Okaia Tephra c. 23 Omataroa T ephra 28.2 R12 Hauparu Tephra 35.87 ? ? R13 Rotoehu Ash 64 R14 R15

5.1 1 Group 3 • Eastern and northeastern Tongariro streams (Fig. 5.6)

Five lahar deposition episodes are recognised in the stream sequences from Oturere to Tahurangi Streams. These are mostly well constrained in age although more unconformities are present in the sequences (Fig. 5.4, 5.6). The oldest deposition episode {T5) consists of two lahar deposits interbedded with fluvial deposits.

111 Ohinepango Stream Wa ihohonu Stream

Ng 1-r--r-r...,....,-1 Tp ��------�

Key to lithology

Figure 5.5 Composite stratigraphic sections for Group 2 streams draining the northeastern flanks of Ruapehu and southeastern flanks of To ngariro. Tephra abbreviations are listed in Ta ble 5.1, lahar deposit nomenclature from text.

112 T5 is preserved beneath a presently unidentified rhyolitic tephra, which is also below an unconformity in the Mangatawai Stream sequence; hence the ages of the T5 units are not well constrained. T4, which commonly consists of several lahar deposits, occurs directly below the Kawakawa Formation, roughly equivalent in age to R10 and R11. Unit T3 comprises lahar deposits that are laterally equivalent to Hinuera Formation fluvial sediments, that are commonly found within the Tongariro streams and are equivalent in age to R09. T2 consists of a distinctive, bright yellow silty matrix debris flow deposit which contains dominantly pebble-sized pumice supported within the matrix and is only recognised in the Oturere, Makahikatoa and Mangatawai Streams. No evidence of lahars equivalent in age to the younger episodes in the Ruapehu streams is recognised in the Tongariro streams; the youngest deposit T1 is equivalent in age to R8 (11 .85-13.08 ka B.P.).

5.12 Lahar distribution on the eastern ring-plain sectors from 23 ka to the present

The areal distribution of each episode of lahar deposition has been mapped from the stream sections and supplementary field data (Fig. 5.7-5.9). The distribution of the oldest units (R12-R1 5, and T5) cannot be mapped because only a few deep sections expose these units. Lahar units of the two oldest mappable episodes have the largest distribution. R 11 and T4 lahar deposits are found over the entire northeastern Ruapehu and eastern Tongariro ring plains (Fig. 5. 7 A). This indicates that active lahar sedimentation was occurring over the entire eastern ring plains before the eruption of the Kawakawa Formation during the early part of the Last Glacial Maximum. Streamflow deposition was also occurring over large areas of the ring plains during this time. R1 0 lahar deposits, however, were derived solely from the northeastern flanks of Ruapehu (Fig. 5. ?B), and are lithologically distinct from the contemporaneous R1 1 and T4 lahars. In the period following the Kawakawa eruption and before the Rerewhakaaitu eruption (i.e. the latter part of the Last Glacial Maximum) much of the area continued to be inundated by lahar and fluvial deposits. Contemporaneous lahar episodes occurred from northeastern Ruapehu sources (R09) and eastern Tongariro sources (T3), coupled with stream sedimentation and reworking of Kawakawa Formation to form Hinuera Formation deposits. In the same time interval, fo rmation of two major landscape features had a major effect on subsequent lahar distribution. The first was a large lava flow which was emplaced between the present Waihohonu and Oturere Streams (Fig 5. ?C) immediately before the eruption of the Kawakawa Tephra. This extensive lava flow served as a divide, effectively isolating lahars with a source in the northeastern Ruapehu and eastern Tongariro catchments. The second landscape feature was deposition of a series of glacial moraines along the Whangaehu and Mangatoetoenui Streams (Fig 5.7C). These moraines blocked the direct drainage from the eastern Ruapehu flanks to a sector of the northeastern Ruapehu ring plain between the present Upper Waikato Stream to the southern tributary of Te Piripiri Stream.

113 Oturere Stream Makahikatoa Stream Mangatawai Stream Mangamate Stream Mangahouhounui Stream Ta uhurangi Stream

Po Mm Wh I I Ra �� I � Wh 2m

__.. __.. � I T1

Key to lithology

Lignite Andesitic lava Andesitic tephras lgnimbrite Pahoka and Pourahu Te phras Bedded sands, silts and gravels Sandy matrix debris-flow deposits Hyperconcentrated-streamflow deposits Silty matrix debris-flow deposits

Figure Composite stratigraphic sections for Group3 streams draining the eastern flanks of To ngariro. Tephra abbreviations listed in 5.6 Ta ble 5.1, lahar deoosit nomenclature from text. A R11 and T4 B R10

0 5km L.LJ 0 5km L.LJ

c D T2

0 5km 0 5km L...J.....J L.LJ

Figure 5.7 Distribution of deposits from lahar episodes: (A) R11/T4, (B) R10, (C) R09/T3 and the Hinuera Formation, (D) T2. Ages of the deposition episodes are given in Fig. 5.4.

115 A ROB and T1 B R07 and R06

0 5km

L.LJ 0 5km L.LJ

C R05 D R04

0 5km 0 5km L.J...... J L.LJ

Figure 5.8 Distribution of deposits from lahar episodes: (A) R08/T1 , (B) RO? and R06, (C) R05, (D) R04. Ages of deposition episodes given in Fig. 5.4.

116 A R03 and R01 B R02

0 5km L.LJ 0 5km L.LJ

Figure 5.9 Distribution of deposits from lahar episodes; (A) R03 and R01 and (B) R02; ages of episodes are given in Fig. 5.4.

117 With the exception of small overflows from the Whangaehu catchment, there have been no lahars in this ring plain sector since. Lahars generated on the eastern Ruapehu flanks since the formation of the moraine ridges have been directed to the southeast into the Whangaehu (Oonoghue 1991) or to the northeast, dominantly into the Mangatoetoenui catchment. All of the lahars after the R09/T3 depositional episode were distributed over much narrower sectors of the ring plains. Their distribution was partially controlled by the lava flow and moraines as previously discussed, together with deepening of drainage channels following the Last Glacial Maximum. T2 episode lahars were distributed only within the Oturere-Mangatawai Stream sector of the Tongariro ring plain (Fig. 5.70). The youngest lahar episode on the eastern Tongariro ring plain, T1 , occurred within the same time range as ROB on the northeastern Ruapehu ring plain. The deposits of these two episodes are confined to areas on either side of the large lava flow between Waihohonu and Oturere Streams (Fig. 5.BA). The distribution of ROB lahars is typical of many of the subsequent deposition episodes on the northeastern Ruapehu ring plain. R07, R06 and R04 lahars all covered the same sector between Waihohonu Stream and the northern tributary of Te Piripiri Stream (Fig. 5.B8, 5.BO). R05 lahar deposits have a narrower distribution within this zone, between the northern tributary of Te Piripiri Stream and Mangatoetoenui Stream (Fig. 5.BC). The three youngest lahar events occurred within the same ring plain sector but were confined to single stream channels - R02 along the northern Te Piripiri Stream tributary, and R03 and R01 within Mangatoetoenui Stream (Fig. 5.9A, 5.98). The youngest lahar to cross the northeastern Ruapehu ring plain was generated on 2B October 1995, and flowed down Mangatoetoenui Stream channel to the Tongariro River (i.e. similar to the path in Fig. 5.9A).

5.13 Source and generation of lahars

A framework for the eruptive activity of Tongariro and Ruapehu volcanoes in the period during which these lahars were generated is contained in Table 5.3. The oldest lahar deposition episode, R15, consists of a thick sequence of debris flow and hyperconcentrated streamflow deposits interbedded with gravelly fluvial sediments derived from Ruapehu volcano. A moderate frequency of large-scale eruptions during this time is reflected by a few layers of andesitic pumice lapilli preserved between the fluvial beds of the sequence. However, very little pumice is found within the lahar and fluvial sediments themselves. In addition, no paleosols are preserved and there is no evidence for major unconformities. These deposits thus appear to represent a period of intermittent volcanic activity with continuous syn- and post-eruptive aggradation on the eastern sector of the Ruapehu ring plain. A lignite deposit within these sediments (Fig. 5.3) contains a grassland and shrubland palynoflora (M.S. McGione 1995, pers. comm.). Above these deposits is a paleosol with the Rotoehu Ash preserved near its base. This enables correlation of the R15 sequence to a widespread period of river aggradation and loess deposition in the lower North Island during the cool climate of marine 818 0 Stage 4

11B (Milne and Smalley, 1979; Pillans, 1988; Cronin et al., 1996(b)). During this period the climate in the region is interpreted from previous palynofloral studies (McGione and Topping, 1983) as having been cool and stormy, with episodes of erosion and widespread cover-bed stripping. The harsher climate would have accelerated physical weathering processes on the volcanoes and thereby provided a ready source of loose sediment on the cone supplementing that produced by eruptions.

Table 5.3 Summary of the eruptive activity of Ruapehu and Tongariro volcanoes for the last 75 ka based on the ring plain tephra record (Topping, 1973; Donoghue et al., 1995; Cronin et al. , 1996(a)). Tongariro volcano Ruapehu volcano Mapping units

0-1.85 frequent, low volume and low frequent (1 per 100 years), low Tufa Trig and magnitude eruptions volume and low magnitude eruptions Ngauruhoe Formations 1.8-2.5 frequent, low volume and low infrequent, low volume and low Mangatawai Tephra magnitude eruptions magnitude eruptions 2.5 - 9.7 frequent, low volume and low infrequent, low volume and low Papakai Formation magnitude eruptions magnitude eruptions 9.7 - 10 very frequent (1 per < 50 years), Mangamate Tephra large volume and large magnitude and Pahoka Tephra eruptions c. 10 large volume and large magnitude Okupata Tephra eruption1 c. 10 1 large volume and large magnitude Pourahu Member of eruption including a pyroclastic flow Bullot Formation c. 10- 22.5 1 large volume and large moderately frequent (1 per 500 Bullot Formation and magnitude eruption years), large volume and large Rotoaira Lapilli magnitude eruptions 22.5 - 35.8 moderate-low frequency (1 per 800 Unnamed years), large volume and large magnitude eruptions. 35.8 - 64 very low frequency (1 per 7 ka), Unnamed large volume and large magnitude eruptions and frequent low volume and low magnitude eruptions. 64 - c. 75 moderately frequent (1 per 600 Unnamed years), large volume and large magnitude eruptions and frequent low volume and low magnitude eru tions.

Redeposition of this loose material then appears to have taken place by lahars or normal streamflow. These were probably triggered by a number of diffe rent mechanisms, such as storms, slope failures, avalanches, glacial collapse and eruptive events. A greater volume of ice and snow on the upper volcano during this cool period may have provided an additional water source for lahar formation. Scouring and melting of glacier ice and snow has been an important mechanism in the formation of lahars in many historical eruptions (Major and Newhall, 1989; Pierson et al. , 1990; Waitt et al., 1994 ) . In addition, at the beginning of the 1995 Ruapehu eruptive sequence, augmentation of lahar volumes by eroded snow from the summit glaciers was an important process (Cronin et al., in press). Deposits of the next three lahar episodes R14, R13 and R12 are interbedded within thick andesitic tephra deposits from infrequent but large-scale eruptions of

119 Ruapehu (Table 5.3), and also contain pumice lapilli within their matrices. R12 deposits in particular contain a 0.2 m-thick andesitic pumice lapilli layer interbedded between debris flow units. R13 and R14 comprise the deposits of single lahar events. All three episodes appear to have accompanied, and were probably caused by sub-plinian tephra eruptions at Ruapehu. The age of these deposits ranges from c. 65-35 ka, during a period of relatively mild and settled climate in the Tongariro region (McGione and Topping, 1983). R1 1 and its stratigraphic equivalent on the Tongariro ring plain, T4, were then deposited before the eruption of the Kawakawa Formation (before 22.6 ka B.P., Wilson et al., 1988). These deposits are thick continuous sequences of debris flow, hyperconcentrated streamflow, and coarse fluvial deposits with occasional andesitic lapilli layers interbedded in the R1 1 deposits. They are widespread over all the examined sectors of both ring plains and indicate that abundant sediment was available on the volcanoes at this time for redeposition. The mechanisms of deposition of these units were likely to be the same as for the R15 lahar deposition episode. The frequency of large-scale eruptions in the interval when R1 1 and T4 were deposited was moderate-low (Table 5.3). However, this interval was the beginning of the greatest period of environmental destabilisation in the North Island of New Zealand (marine 818 0 stage 2, or the Last Glacial Maximum - 23-13 ka) over the last c. 120 ka (Pillans et al., 1993). R1 0 is the deposit of a single event or a closely spaced series of lahar pulses within the same time period as R1 1 and T4 . However, this unit is lithologically distinct from them. lt has a clay-rich silty matrix and hydrothermally altered clasts, indicating that it has been derived from a collapse of a hydrothermally altered section of the former Ruapehu cone. The R09/T3 lahar deposition episode represents lahar aggradation during the Last Glacial Maximum (marine 8180 stage 2) after the Kawakawa eruption. lt is coeval with deposits of redeposited Kawakawa ignimbrite mixed with andesitic sand and gravels, which have been previously mapped as Hinuera Formation in the northern sector of Tongariro ring plain (Topping and Kahn, 1973). R09/T3 deposits are also widespread over much of the ring plain examined. They appear to have ceased accumulating immediately before the eruption of the Rerewhakaaitu Tephra. This is coincident with climatic conditions beginning to ameliorate in the Tongariro region (McGione and Topping, 1977). The frequency of large scale eruptions during this period was moderate (Table 5.3), although this activity continued after widespread lahar deposition had ceased. Deposits of the subsequent lahar episodes are distributed over much narrower sectors of the two ring plains. This is a result of both landform changes, as discussed previously, as well as seemingly reduced availability of loose sediment on the flanks of the two volcanoes with climatically induced landscape stability after the Last Glacial Maximum. Where lahars continued to be deposited they appear to have been more channellised than previously. This is probably due to incision of stream channels following the Last Glacial Maximum when more water and less sediment became available. On the eastern Tongariro ring plain, the Oturere-Mangatawai Stream sector

120 was the only area active following the Last Glacial Maximum where the deposits of two lahar episodes, T1 and T2, are preserved. T2 deposits are rich in andesitic pumice (up to 30 % by volume) and appear to be associated with the eruption of Rotoaira Tephra from Tongariro volcano at 13.8 ka (Topping, 1973). T1 , the youngest lahar deposit on the Tongariro ring plain, is within the same time frame as the first of eight lahar episodes on the northeastern Ruapehu ring plain, restricted between the main tributary of the Te Piripiri and Waihohonu Streams. During the time of the lahar deposition episodes R08/T1 to R02 (c. 13-9 ka) both volcanoes were erupting thick sub-plinian tephra layers with a high frequency, including the Mangamate and Pahoka Tephras from Tongariro (Topping, 1973) and the younger Bullot Formation tephras from Ruapehu (Donoghue et al., 1995). Several of the lahar deposition episodes appear to be directly associated with these eruptives. The most notable are: R06 with the Pourahu Member of the Bullot Formation, R05 with Pahoka Tephra, R04 with Oturere Lapilli member of the Mangamate Tephra, R03 with Ohinepango/Waihohonu Tephra members of the Mangamate Tephra, and R02 with the Poutu Lapilli member of the Mangamate Tephra. The association with explosive volcanism provides a triggering mechanism and a large volume of source material for these lahar episodes but it does not completely explain their consistently restricted area of deposition. The Waihohonu and Oturere Streams on the Tongariro ring plain and the Mangatoetoenui Stream on the Ruapehu ring plain all have evidence of glacial moraines in their headwaters (Mathews, 1967; Hackett, 1985; McArthur and Shephard, 1990). None of the other streams examined in this study do. The Waiohau Tephra (1 1 .85 ka B.P.) occurs near the base of the coverbed sequence on the moraines beside Waihohonu Stream, the Rerewhakaaitu Tephra (14.7 ka B.P.) on the Oturere Stream moraines, and the Pourahu Tephra (c. 10 ka) on Mangatoetoenui Stream moraines. This suggests the glacier in the fo rmer Oturere valley retreated before the glaciers in the Waihohonu and Mangatoetoenui valleys. This retreat is also reflected by the cessation of frequent lahars in the Oturere Stream after T1 at c. 14 ka, in the upper Waihohonu Stream (upstream of the confluence with Ohinepango Stream) from c. 15 ka, and the Mangatoetoenui Stream (where a glacier still exists on the upper volcano slopes) at c. 9ka. The presence of three retreating glaciers at these times provides a potential source of snow and water for lahar formation when eruptive products landed on them. Also moraine deposits are a ready source of sediment. Thus, the lahar record suggests these catchments were predisposed to lahar formation compared to other catchments without glaciers. Furthermore, units such as the Pourahu Member of the Bullot Formation had an associated pyroclastic flow (Donoghue et al. , 1995), which could have melted a large amount of snow and ice to form the lahars of the R06 episode. The youngest lahar deposition episode occurred between 5.3 and 5.2 ka B.P. in the Mangatoetoenui Stream when the climate was wetter and milder than the present (McGione and Topping, 1977). The activity of the two andesite volcanoes had reduced in magnitude at this time, with fine ash of the Papakai Formation accumulating slowly on the

121 ring plain. This single lahar event could have had many triggering mechanisms, the most likely being an eruption which was not large enough to be represented individually in the tephra record. The 1995 lahar in the Mangatoetoenui Stream was a dominantly monolithologic unit derived from the sudden rainfall-induced failure of large thicknesses of tephra (up to 80 cm thick) erupted onto the Mangatoetoenui Glacier.

5.14 Summaryand Conclusions

Using the coverbed stratigraphy of rhyolitic and andesitic tephra, 15 lahar episodes are recognised in the northeastern Ruapehu ring-plain record, ranging in age from >65 to 5 ka. Five episodes are recognised on the eastern Tongariro ring plain ranging in age from >23 to 14 ka. The two main periods of lahar and fluvial aggradation were from c. 75-65 ka and 23 to 14 ka, corresponding to the two coolest periods of the Last Glacial (marine 8180 Stages 4 and 2). The cool and stormy climate of these times coupled with an expanded area of glaciers appeared to result in greater sediment supply for lahar formation. Greater volumes of snow and ice also increased the availability of water for lahar generation. The triggering mechanisms for these lahars could have been eruptive activity, storm events, slope failures, avalanches and glacier collapses. Two events during the period c. 23-14.7 ka B.P. strongly influenced the distribution of subsequent lahars. Firstly, the eastern Tongariro and Ruapehu ring plains were separated by a large lava flow emplaced between the present Waihohonu and Oturere Streams immediately before 22.5 ka. Secondly, Last Glacial Maximum moraines along the upper Whangaehu and Mangatoetoenui valleys blocked lahars from the upper eastern flanks of Ruapehu to the ring-plain sector between the Upper Waikato and Te Piripiri Streams. Subsequent lahars generated on the eastern Ruapehu flanks have thus been channelled to the southeast down the Whangaehu catchment and northeast into mostly the Mangatoetoenui catchment. Many of the late glacial and postglacial lahar episodes appear to be related to plinian and sub-plinian tephra eruptions of Tongariro and Ruapehu volcanoes. However, both tephra deposition and water supply were required to form lahars. lt appears major lahars at this time were formed only where tephra fell onto catchments with valley glaciers. The cessation of lahars on the eastern flanks of Tongariro before 11.6 ka B.P. coincided with vegetational stabilisation of the volcanic slopes and glacial moraines. On the northeastern Ruapehu ring plain, lahar deposition continued until 5.2 ka B.P., possibly because the glaciers were slower to retreat, particularly in the Mangatoetoenui catchment.

5.15 Acknowledgements We gratefully acknowledge funding from the New Zealand Vice-Chancellor's Committee,

122 Massey University Graduate Research Fund, the Helen E. Akers Scholarship Fund, and the Department of Soil Science of Massey University. We also thank A.S. Palmer, R.B. Stewart and J.A. Lecointre for their useful reviews of an earlier version of the manuscript, and B. F. Houghton and A.W. Walton for their comprehensive and useful reviews.

5.16 References

Cronin, S.J., Neall, V.E. and Palmer, A.S., 1996(b). Investigation of an aggrading paleosol developed into andesitic ring plain deposits, Ruapehu volcano, New Zealand. Geoderma, 69: 119-135. Cronin, S.J., Neall, V.E., Stewart, R.B. and Palmer, A.S., 1996(a). A multiple parameter approach to andesitic tephra correlation, Ruapehu volcano, New Zealand. J. Volcano!. and Geotherm. Res. 72, 199-215. Cronin, S.J., Neall, V.E., Lecointre, J.A., Palmer, A.S., in press. Unusual "snow slurry" lahars from Ruapehu volcano, New Zealand, September 1995. Geology. Donoghue, S.L., 1991 . Late Quaternary volcanic stratigraphy of the south-eastern sector of Mount Ruapehu ring plain, New Zealand. Unpub. PhD thesis, Massey University, Palmerston North, New Zealand. Donoghue, S.L., Neall, V.E. and Palmer, A.S., 1995. Stratigraphy and chronology of late Quaternary andesitic tephra deposits, Tongariro Volcanic Centre, New Zealand. J. Roy. Soc. N. Z., 25: 115-206. Froggatt, P.C. and Lowe, D.J., 1990. A review of late Quaternary silicic and some other tephra formations from New Zealand: their stratigraphy, nomenclature, distribution, volume, and age. N. Z. J. Geol. and Geophys., 33: 89-109. Gregg, D.R. , 1960. The geology of Tongariro Subdivision. N. Z. Geol. Survey Bull., 40. Hackett, W.R., 1985. The geology and petrology of Ruapehu volcano and related vents. Unpub. PhD Thesis, Victoria University of Wellington, New Zealand. Hackett, W.R. and Houghton, B.F., 1989. A facies model for a Quaternary andesitic composite volcano: Ruapehu, New Zealand. Bull. Volcano!., 51: 51-68. Major, J.J. and Newhall, C.G., 1989. Snow and ice perturbation during historical volcanic eruptions and the formation of lahars and floods; a global review. Bull. Volcano!., 52: 1-27. Mathews, W.H., 1967. A contribution to the geology of the Mount Tongariro massif, North Island, New Zealand. N. Z. J. Geol. and Geophys., 10: 1027-1038. McArthur, J.L. and Shepherd, M.J., 1990. Late Quaternary glaciation of Mt. Ruapehu, North Island, New Zealand. J. Roy. Soc. N. Z., 20: 287-296. McGione, M.S. and Topping, W.W., 1977. Aranuian (post-glacial) pollen diagrams from the Tongariro region, New Zealand. N. Z. J. of Botany, 15: 749-760. McGione, M.S. and Topping, W.W., 1983. Late Quaternaryveget ation, Tongariro region, central North Island, New Zealand. N. Z. J .. Botany, 21: 53-76. Milne, J.D.G. and Smalley I.J., 1979. Loess deposits in the southern part of the North Island of New Zealand: an outline stratigraphy. Acta Geologica Academiae

123 Scientarium Hungaricae, 22: 197-204. Nairn, I.A., Wood, C.P and Hewson C.A.Y., 1979. Phreatic eruptions of Ruapehu: April 1975. N. Z. J. Geol. and Geophys., 22: 155-173. Palmer, B.A., Purves, A.M. and Donoghue, S.L., 1993. Controls on the accumulation of a volcaniclastic fa n, Ruapehu composite volcano, New Zealand. Bull. Volcano!., 55: 176-189. Pierson, T.C. and Costa, J.E., 1987., A rheologic classification of subaerial sediment water flows. G.S.A. Reviews in Engineering Geology, VII: 1-12. Pierson, T.C., Janda, R.J., Thouret, J.-C. and Borrero, C.A., 1990. Perturbation and melting of snow and ice by the 13 November 1985 eruption of Nevado del Ruiz, Colombia, and consequent mobilisation, flow and deposition of lahars. J. Volcano!. and Geotherm. Res., 41 : 17-66. Pillans, B.J., 1988. Loess chronology in Wanganui Basin, New Zealand. In: Eden, D.N. and Furkert, R.J. (eds) Loess - Its Distribution Geology and Soils: 175-1 91 . A.A. Balkema, Rotterdam. Pillans, B., McGione, M., Palmer, A., Mildenhall, D., Alloway, B., Berger, G., 1993. The Last Glacial Maximum in central and southern North Island: a paleoenvironmental reconstruction using the Kawakawa tephra fo rmation as a chronostratigraphic marker. Pal., Pal., Pal., 101: 283-304. Scott, K.M., 1988. Origins, behaviour, and sedimentology of lahars and lahar-runout flows in the Toutle-Cowlitz River system. U.S.G.S. Prof. Pap. 1447-A: 76pp. Smith, G.A., 1986. Coarse-grained nonmarine volcaniclastic sediment: Terminology and deposition process. G.S.A. Bull., 97: 1-10. Smith, G.A. Fritz, W.J., 1989. Volcanic influences on terrestrial sedimentation. Geology 17: 375-376. Topping, W.W., 1973. Tephrostratigraphy and chronology of late Quaternary eruptives from the Tongariro Volcanic Centre, New Zealand. N. Z. J. Geol. and Geophys., 16: 397-423. Topping, W.W. and Kohn B.P., 1973. Rhyolitic tephra marker beds in the Tongariro area, North Island, New Zealand. N. Z. J. Geol. and Geophys., 16: 375-395. Waitt, R.B., Gardner, C.A., Pierson, T.C., Major, J.J., Neal, C.A., 1994. Unusual ice diamicts emplaced during the December 15, 1989 eruption of Redoubt Volcano, Alaska. J. of Volcano!. and Geotherm Res. 62: 409-428. Wilson, C.J.N., Switzur, RV. and Ward, A.P., 1988. A new 14C age for the Oruanui (Wairakei) eruption, New Zealand. Geol. Mag.,125: 297-300. Wilson, C.J.N., Houghton, B.F., Lanphere, M.A. and Weaver, S.D., 1992. A new radiometric age estimate for the Rotoehu Ash from Mayor Island volcano, New Zealand. N. Z. J. Geol. and Geophys., 35: 371-374. Wilson, C.J.N., 1993. Stratigraphy, chronology, styles and dynamics of late Quaternary eruptions from Taupo Volcano, New Zealand. Phil. Trans. Roy. Soc. of London A, 343: 205-306.

124 CHAPTER 6: LAHAR HISTORY AND LAHAR HAZARD OF THE TONGARIRO RIVER

6.1 Introduction

The Tongariro River drains a large portion of the northeastern Tongariro Volcanic Centre and thus has the potential of being a conduit for devastating lahars as has happened in the geological past. Built beside the river, Turangi, a town of more than 4 000 people is potentially at risk from these lahars. The aim of this chapter is also one of the main

objectives of the overall study outlined in Chapter 1 - to establish what is the hazard to Turangi and the surrounding area from lahars in the Tongariro River. This part of the study relies on many of the developments made in previous chapters. Using the rhyolitic and andesitic tephrochronology established for the study area in Chapters 2 and 3, the lahar stratigraphy along the Tongariro River was investigated in the same manner as for Chapter 5. By combining the Tongariro River stratigraphy with that developed for the northeastern Ruapehu and eastern Tongariro ring plains (Chapter 5), the lahar hazards of the entire Tongariro River catchment have been assessed. This part of the study has been written as a manuscript which has been submitted to the New Zealand Journal of Geology and Geophysics. The manuscript has multiple authors and the contributions of the authors were as follows:

S. J. Cronin: Principal investigator Carried out all: Field descriptions and sampling Tephra and lahar correlation and mapping Drawing and designation of zones on the hazard map Preparation and writing of the manuscript

V. E. Neall A.S. Palmer: Advisers Aided the study by: Discussion of methodology and lahar hazard designation Editing and discussion of the manuscript and lahar hazard map

125 6.2 LAHAR HISTORY AND LAHAR HAZARD OF THE TONGARIRO RIVER, NORTHEASTERN TONGARIRO VOLCANIC CENTRE, NEW ZEALAND

Shane J. Cronin, V.E. Neall and A.S. Palmer Department of Soil Science, Massey University, Private Bag 11 222, Palmerston North, New Zealand.

Submitted to: New Zealand Journalof Geology and Geophysics

Abstract Laharic surfaces beside the Tongariro River have been mapped and dated using andesitic and rhyolitic marker tephras. Coupling the stratigraphic record obtained with that of the Tongariro and Ruapehu ring plains, has enabled a history of lahars occurring in the Tongariro River to be compiled. This has formed the basis of a lahar hazard map for the entire catchment. Eight lahar hazard zones with assigned recurrence intervals ranging from 1 in 35 years to 1 in >15 000 years have been mapped. Lahar surfaces between the ages of 14.7 and 9.8 ka B.P. cover the greatest areas, while younger lahar surfaces are confined to lower and restricted surfaces closer to the present river channel. Holocene lahar deposits along the Tongariro River are not as well preserved as older units, probably because the Holocene lahars were confined to a more deeply incised channel where their deposits were more readily eroded following emplacement. All recorded lahars in the Tongariro catchment post-1 1 .85 ka B.P. have been derived from Ruapehu volcano. The Mangatoetoenui Stream has been the conduit for the greatest number of Holocene lahars and for all of the historic ones. Most of Turangi is built on a surface which has not been inundated by a lahar since c. 10 ka B.P. However, the area identified where high risk affects the greatest population is part of Turangi, having been inundated by a lahar since 1850 years B.P. The infrastructure at greatest risk includes the State Highway 1 bridge across the Mangatoetoenui Stream and the Rangipo Dam and power station.

6.3 Keywords

Volcanic hazards, volcanic risk, lahar hazard map, lahars, Ruapehu volcano, Tongariro volcano, Tongariro River, Late Quaternary lahars, Holocene Lahars.

6.4 Introduction

The Tongariro catchment drains the northeastern portion of the Tongariro Volcanic Centre and includes the northeastern flanks of Ruapehu volcano and the eastern and northeastern flanks of Tongariro volcano. Lahars can enter the Tongariro River via a large number of tributary channels draining these two volcanoes. The most recent example was a small lahar from Ruapehu volcano in October 1995.

126 At risk from lahars along the Tongariro River is the town of Turangi with a population of 4239 in 1991 (Department of Statistics, 1992). Two smaller settlements, Rangipo and Tokaanu and the Rangipo Prison lie farther from the river, but are also located on laharic and fluvial surfaces. The Tongariro Power Development which utilises both the Tongariro river channel and water from its tributaries is also at risk from lahars. In 1995 laharic and fluvial sediment filled a large portion of the Rangipo dam, sited within the Tongariro River channel. This and additional sediment over the following few months led to excessive wear of turbine blades in the Rangipo Power Station. Also at risk from lahars along the Tongariro River is tourist revenue from trout fishing and other recreational activities. To assess the lahar hazards within the Tongariro River and its surrounds we have mapped the laharic surfaces and sediments along the River and dated them using rhyolitic and andesitic cover- and interbeds. Combining this with a knowledge of the lahar stratigraphy on the northeastern Ruapehu and eastern Tongariro ring plains (Cronin and Neall, in press) we have produced an integrated lahar hazard map for the entire Tongariro catchment.

6.5 Setting

Tongariro and Ruapehu are the two largest and most recently active volcanoes of the Tongariro Volcanic Centre (Gregg, 1960). Ruapehu was last active in 1996. lt is the highest peak in the North Island at 2797 m and consists of a 110 km3 composite cone with a surrounding ring plain of similar volume, comprising mainly volcaniclastic deposits (Hackett and Houghton, 1989). Tongariro volcano is a slightly smaller massif made up of several coalescing volcanic cones, the largest of which is the recently active cone of Ngauruhoe (Mathews, 1967; Nairn and Self, 1978). Tongariro volcano is also surrounded by an extensive ring plain. The ring plain deposits of these two volcanoes are confined in the east by the Kaimanawa Mountains and extend down drainage channels to the north and south. The Upper Waikato Stream is the southern boundary of the volcanic portion of the Tongariro catchment (Fig. 6.1 ). South of this boundary (a low divide on the Whangaehu fan), drainage from the flanks of Ruapehu flows southwards into the Whangaehu River. From its origins at the junction of the Waipakihi River and Upper Waikato Stream, the Tongariro River flows for 47 km before entering Lake Taupo. Its gradient ranges from 1 m drop in 43 m to 1 m in 225 m, before it becomes meandering within a delta formed into Lake Taupo. Along its path, 9 major and several other smaller streams rising from the flanks of the Ruapehu and Tongariro volcanoes join the River.

6.6 Terminology

The word lahar is here used to describe, "a rapidly flowing mixture of rock debris and water (other than normal streamflow) from a volcano" (Smith and Fritz, 1989).

127 Lake Ta upo

Taupo Volcanic Zone

Pihanga A

Mt To ngariro .A.

MI N uh) � � 1500 m

39° 10 I

- State Highways 0 5km

Figure 6.1 Location of the To ngariro catchment, draining the northeastern To ngariro Volcanic Centre in the central North Island of New Zealand. The nine major tributary streams which drain from Ruapehu and To ngariro volcanoes into the To ngariro River are labelled.

128 Lahars are generally subdivided on the basis of differing sediment concentrations and resulting differences in rheology. A continuum of flow behaviour and rheology occurs between normal Newtonian streamflow and plastic plug-flows (Pierson and Costa, 1987). Most lahars fall into two main rheologic categories which have been defined on the basis of field and laboratory experiment observations; debris flows and hyperconcentrated streamflows. Hyperconcentrated streamflows are defined as containing 20-60% sediment by volume (Beverage and Culbertson, 1964 ). More importantly however, the onset of hyperconcentrated streamflow behaviour is when a sediment-water mixture begins to attain a measurable yield strength (i.e. it becomes a non-Newtonian fluid). Sediment is supported within these flows dominantly by clast-clast collisions/interactions and turbulence, and deposition is by a rapid grain by grain settling process at the base of the flow (Pierson and Scott, 1985; Smith, 1986). Deposits typical of hyperconcentrated flows are usually sand dominated, horizontally bedded, poorly sorted and clast supported. When the sediment concentration in the flow approaches 60% (the actual content varies according to the particle-size distribution of the sediment), yield strengths of the flow increase markedly. At this stage the flow begins to behave like a laminar flowing, single-phase plug rather than sediment suspended in water (Pierson and Costa, 1987). Clast support in these flows is by matrix support and buoyancy, and deposition is en masse, the deposits representing a frozen state of the flow. Debris flow deposits are dominantly matrix supported, massive and very poorly sorted (Smith, 1986).

6. 7 Hazards of Lahars

The hazards posed by lahars have been vividly portrayed by a number of disasters throughout the world in recent history. In 1953, in a local example, 151 people were killed when their train plunged into the Whangaehu River as a lahar undermined the rail bridge (Stillwell et al., 1954). In 1985, the town of Armero in Colombia and 23 000 of its inhabitants were inundated by lahars from Nevada del Ruiz (Pierson et al., 1990; Voight, 1990). In 1991 and following years, lahars in rivers radiating outward from Mt. Pinatubo in the Philippines caused at least 143 fatalities. They destroyed bridges and inundated towns, roads and huge areas of arable land, forcing the evacuation of more than 70 000 people (Philippine Institute of Volcanology and Seismology, 1992; Pierson et al. , 1992). Hazard to life from lahars results from their high speeds and sediment concentrations. People and animals are swept away or entombed in sediment. Bridges and other man-made structures are also easily destroyed by large, highly competent flows. Lahars can also deposit sediment across large areas of productive land, not only destroying crops but potentially resulting in a permanent loss of production. Other hazards can result from the acidity, toxicity and heat of some types of lahars.

129 6.8 Methods

The surfaces and sediments in this study have been dated using andesitic and rhyolitic tephra cover- and interbeds. Lahar surfaces were mapped from field exposures and using aerial photographs. Eighteen dated rhyolitic tephras erupted from Taupo Volcanic Zone calderas are preserved within the Ruapehu and Tongariro ring plain and Tongariro River sequences (Topping and Kohn, 1973; Donoghue et al., 1995; Cronin and Neall, in press). These tephras were identified by their stratigraphic positions, physical appearances, mineralogy and glass chemistry (Cronin et al. in press(a)). In addition, 11 dated andesitic tephras erupted from Tongariro and Ruapehu volcanoes occur within the sequences (Topping, 1973; Donoghue et al. , 1995). These and further andesitic marker tephras were identified by a similar range of methods as those used for the rhyolitic tephras (Cronin et al. , 1996). All the marker tephras discussed in this study are listed in Table 5.1 (Chapter 5). Laharic deposits within the stratigraphic sequences were interbedded with fluvial sediments, tephras and lava flows. The laharic deposits were distinguished from fluvial sediments by their laterally continuous and characteristic sedimentology outlined previously. In constructing the lahar hazard map, laharic surfaces of differing ages were delineated into zones and a lahar recurrence interval was assigned to each zone. For the three youngest zones this was a simple process of calculating the ratio of the number of lahars in a given time, e.g. within the historic record (1 00-150 years), or since the Taupo Tephra (1850 years B.P.; Froggatt and Lowe, 1990). For older lahar surfaces this approach yields results which are not considered valid, e.g. a surface mapped in this study has not been inundated by a lahar since c. 10 ka B.P. This surface comprises the deposits of several lahars, and using tephra interbeds various recurrence intervals can be calculated. Since 13.8 ka B.P. the recurrence interval is 1 in 2 800 years, and since 22.6 ka B.P. it is 1 in 3 200 years. These intervals are not representative of a surface which has not been crossed by a lahar in the last 10 ka, nor are they representative of the actual frequency of the lahar deposition (from 22.5 to 13.8 ka B.P.=1 in 4 400 years, and from 13.8 to 10 ka B.P.=1 in 800 years). In situations such as this, the approach in this study was to assign a recurrence interval based on the last time a given surface was inundated by a lahar.

6.9 Lahar history of the Tongariro River

6.9.1 Stratigraphic record within the tributary streams

The number and timing of lahars entering the Tongariro River may be obtained from the lahar record preserved within its volcanic sourced tributaries (Cronin and Neall, in press). This indicates that prior to the eruption of the Rerewhakaaitu Tephra (14.7 Ka B.P.), multiple lahars occurred within almost all of its major tributaries (Table 6.2). In contrast,

130 Holocene lahars are restricted to three streams on the northeastern Ruapehu ring plain. The Mangatoetoenui Stream has contributed the largest number of post-glacial lahars to the Tongariro River. During the 1995/96 eruption episode of Ruapehu volcano, the only lahars to enter the Tongariro River travelled down the Mangatoetoenui Stream.

6.9.2 Stratigraphic record within the Tongariro River valley

6.9.2A Highest surface The oldest exposed sequences in the Tongariro River valley form the highest surface, which is commonly 15-30 m above the present river level. This surface is extensive throughout the catchment between Upper Waikato Stream and the town of Turangi. Sections through it are described in exposures at various locations along the Tongariro River (Fig. 6.2). Close to the Tongariro River the surface has been eroded and subsequently been infilled by the Taupo lgnimbrite. As a result of this infilling, lahar surfaces of differing ages have been buried to form a single younger surface. This complexity is exemplified by the diffe rence in age of the lahars comprising the highest surface described at Site 1 with those at Sites 2 and 3 which have been overlain by 4 to 6 m of pumiceous ignimbrite to form a laterally continuous surfa ce. At Sites 2 and 3 beneath the pumice veneer are formerly lower and younger laharic surfaces than at Site 1. Sites 2 to 7 from 1.5 to 36 km down the river all have coverbed sequences of similar ages. At Site 2, the lahar deposits above an unconformity are younger than any recorded in the Upper Waikato Stream and are probably derived from the Te Piripiri catchment. Sites 2 to 5 all record the deposits of several lahars below either the Pourahu or the Pahoka Tephras (c. 9.8-10 ka B.P.). At Sites 6 and 7 the Pahoka and Pourahu tephras have thinned beyond recognition, but the lower members of Mangamate Tephra and the Poronui Tephra are present, providing a minimum age of c. 9.7-9.8 ka B.P. for the underlying lahar deposits. None of the many lahars younger than 10 ka B.P. that are recorded by deposits in the Te Piripiri, Mangatoetoenui or Waihohonu Streams (Table 6.2) are represented in any of the sections described through this high surface. The Rotoaira Tephra preserved at Site 6 provides a lower age limit for a package of several lahar deposits. This indicates that this lahar package was probably derived from the Te Piripiri, Mangatoetoenui, Waihohonu, Oturere, and Makahikatoa catchments. Lahar deposits preserved below the Kawakawa Tephra at Sites 6 and 7 could have been derived from lahars in any of the tributary streams (Table 6.2). The post-Kawakawa lahars described in the Tongariro valley sequences cannot be individually correlated downstream on the basis of lithology or stratigraphy. However, consistent changes in the sedimentology of the overall package of post-22.6 ka B.P. lahar deposits are observed. In the upstream sites most of the deposits are: coarse grained - containing pebbles-large boulders, matrix-supported with a sandy or silty matrix, poorly sorted and massive.

131 Table 6.2 Number and timing of lahars in tributary catchments of the Tongariro River, based on the exposed stratigraphic record in each catchment (Cronin & Neall in press).

Marker horizons age (ka B.P.) Upper Waikato Te Piripi ri Manga toetoenliL_ Waihohonu Oturere Makahikatoa Manga tawai Mangamate Puketarata Mangahouhounui

1995 AD I n 2 I Hinemaiaia Tephra 5.2-3.95 I 1 I Motutere Tephra 5.8-5.3 I 1 I Mm, Poutu Lapilli c. 9.7 Mm, Wharepu Tephra c. 9.7 I 1 I Mm, Ohinepango Tephra c. 9.9 Mm, Waihohonu Lapilli c. 9.9 1 1 1 Mm, Oturere Lapilli c. 9.9 Karapiti Tephra 10.1 I 1 1 I Pahoka Tephra c. 10 1 1 1 Bt, Pourahu Member c. 10 1 1 1 Waiohau Tephra 11.85 r- -1-·---.,- I I 1 1 I Rotorua T ephra 13.08 Rotoaira Tephra 13.8 I 2 1 1 I Rerewhakaaitu Tephra 14.7 I several several several several several 1+ 1+ I 1+ 1+ I Kawakawa Tephra 22.59 I I I I several 1+ I I several several several several several I several Okaia Tephra c. 23 Omataroa Tephra 28.2 I 2 HauparuTephra 35.87 Unexposed Unexposed I 1 RotoehuAsh 64 I several

132 ...... w w

, '..(, _ ',' ; - �� - ::·. .� - �,� Ra . - � .. ' , I k ._ : ' - o-.o .-:- " ·.o ·.-:- · ·�:· ��.--�: o �' . .. �- · ' , ..t '. 2 m - . �=-,?_: �=-�. :-; , �� ...: · · . ..; _ . .�: �"..--\ . � - . - :- 0-:- ;.-..; o· ' . � - � -· •. · - ._ . _. ,:_ - . . "'i -- . - �-. · � ':_/":-_0 ° - _0 , . _ o . _-: · . o · .-: ·i :..:.- o - -.:- o- - . · o- -. · . · Q · - · o · - ·: -� -.·-: �,/� >;,, _ o '� - ....: -.:..: _:_ c. ��: . � . . . - - . o .�-;.�:� -1 , - -.: ·o·.o·:-:- · - . _ o--- o· : '"� � . . . -...'�� ' 0"':"" ;.·-: _o-:- ;. • ,- .. 5m · , I ' , · ,_ I , · 2m �, , , , (' -. ' ' . ,, , ...- ' , ,_ - , 2m , ' , _ . .� ' "'..( ''.' ... , -, - "�"; , , I ' Key to lithology .- .., \ '' I Andesitic lava I Sandy matrix debris flow deposits lgnimbrite Pahoka and Pourahu Te phras Hyperconcentrated streamflow deposits Andesitic tephras Silty matrix debris flow deposits Bedded sands, silts and gravels � Lignite u = Unconformity

Figure Stratigraphy of sections preserved through the highest laharic surface beside the To ngariro River. Sites 1-7 are labelled on Figs. 4 and 5. 6.2 Sites 2-7 are located at the labelled distances downstream of the start of the To ngariro River (at the junction of the Waipakihi River and the Uooer Waikato Stream). In contrast in downstream sites the deposits are: dominated by sands and large clast poor; clast-supported; weakly planar or horizontally bedded with rare larger clasts in horizontal "strings". These changes in sedimentology indicate changes in the rheology of many of the lahars as they travelled downstream. In upstream localities, the deposits indicate the lahars were likely to be debris flows, depositing sediment en masse. The downstream deposits imply finer-grained, more dilute hyperconcentrated streamflows, with incremental deposition of many thin layers to build up each deposit. These lahars appear to have transformed from debris flow to hyperconcentrated streamflow between c. 25 and 29 km along the Tongariro River. This was probably due to loss of sediment from the flows by deposition and their dilution by incorporating river water in their paths. Similar lahar transformations have been described by Pierson and Scott (1985) at Mt. St. Helens, and Cronin et al. (in press(b)) in 1995 lahars in the Whangaehu River.

6.9.28 Lower surfaces Small areas of low surfaces are preserved close to the main channel in the upper reaches of the river. These are flood plains, composed of recent (post-1850 years B.P.), coarse (bouldery and cobbly) alluvial deposits, mostly with no stratigraphy exposed. In the lower reaches from the town of Turangi northwards, the lower surfaces are extensive. Where the surfaces are more extensive, only those preserving a stratigraphy were described. In all described exposures, lahar stratigraphy is very difficult to interpret because of extensive fluvial reworking of lahar deposits and poor preservation of tephra marker beds. However, a sequence of at least four lahar deposits is preserved in a small number of exposures (Fig. 6.3). At Site A (Fig. 6.3) two pre-Taupo Tephra lahar deposits are preserved, with a variable thickness of fluvial sands between them. Deposit 2 is unconsolidated and finer grained than the large boulder-bearing, and firmly compacted deposit 3. Site A has been inundated by the River since the deposition of Taupo Pumice, but this deposit is not laharic. At Site B a similar stratigraphy is preserved, and the sedimentologies of lahar deposits 2 and 3 are also similar to Site A. Lahar deposit 2 was probably emplaced soon before the Taupo lgnimbrite, because little or no reworking or soil development is evident at the top of the lahar deposit. Site B also preserves the deposit of at least one lahar beneath lahar deposit 3. Downstream, at Site C the lahar deposit preserved at the base of the sequence is correlated to deposit 3 from its grainsize characteristics, sedimentology and firmly compacted nature. The Hinemaiaia Tephra (Hm on Fig. 6.3) provides a minimum age for this lahar. Lahar deposit 2 is not preserved at this site. At Sites D and E, which are on lower surfaces than the first three localities, the basal lahar deposits are correlated to lahar deposit 2, because of their unconsolidated nature and finer grainsize than deposit 3. Large parts of the upper portions of lahar deposit 2 at these sites and many others are reworked, the fine grained matrix having been elutriated leaving a cobbly-bouldery matrix-poor clast-supported deposit.

134 Site B Site A Stag pool Sand Pool Site C 8m Hydro Pool Site Tp G Swirl Pool 7 Site E +Tp �� Judges Pool 6

...... 5 Site D ....>. Cableway . . Site H VJ Upper Island 4 (]1 Site F Pool SH1 bridge 3

1 +Tp

·�------· � River Level + Tp Includes reworked Ta upo Te phra clasts d ::��=� :�;ds E22Z2J Ta upo lgnimbrite 1 1 1 1 1 1 Soil development m ��;s��s�;�:�::�:::e�: an��elly sands r.::>::::::::::::l � Hinemaia Te phra I I Andesitic tephra Figure 6.3 Stratigraphy of sections through the lower laharic surfaces alongside the To ngariro River. Sites A to G are located on Figs. 4 and 5. Sites F and H exemplify many of the lowest surface exposures where only bedded sands and clast-supported cobbles and boulders are preserved. In these sites the lahar deposits appear to have been completely eroded or reworked. At site G the youngest (pre-1995) lahar deposit in the Tongariro catchment is preserved (lahar deposit 1 ). This deposit contains reworked Taupo Tephra clasts, indicating a post-1850 year B.P. age. Alluvial sands and silts with two episodes of weak soil development occur above the lahar deposit providing no definite minimum age. Lahar deposit 2 is also preserved at Site G, and in many places its upper portion is reworked. The deposit 4 lahar was probably derived from either Te Piripiri or Mangatoetoenui Streams, based on the presence of lahar deposits in a similar stratigraphic position (Table 6.2 and Cronin and Neall, in press). The lahar that emplaced deposit 3 was probably derived from the Mangatoetoenui Stream, where a lahar deposit in the same stratigraphic position has been described. Lahar deposits 2 and 1 are not preserved in upstream areas, thus it is difficult to assess from which catchments they were derived. However, in the late Holocene many lahars were derived from Ruapehu volcano in the Whangaehu catchment (the Onetapu Formation, Donoghue, 1991 ; Hodgson, 1993; Palmer et al., 1993). So deposits 2 and 1 are likely to be from Ruapehu, and given the 1995 example, the Mangatoetoenui Stream was their probable conduit. The fact that the youngest lahar deposits are not well preserved in the Tongariro "'t:. catchment can be explained by comparison to the 1995 lahars in the Whangaehu catchment (Cronin et al., in press(b)). Of 35 lahars recorded in the Whangaehu River from September to December 1995, one year later only a few localities preserve the deposits from only one of these - the largest. However, large amounts of reworked deposits occur. The reason for reworking of the 1995 lahar deposits is that they were mostly confined to the river channel, so streamflow remobilised them immediately following their deposition. The same circumstances probably applied to lahar deposits within the Tongariro River in the late Holocene when the channel was incised almost as deeply as at present. In contrast, during the late glacial and early post-glacial period, the Tongariro River was probably not confined to a deeply incised channel. This resulted in emplacement of lahar deposits across a broad area, now represented by the highest lahar surface preserved extensively throughout the catchment.

6.10 Lahar hazards in the Tongariro River

The lahar hazard map of the Tongariro catchment (Figs. 6.4 and 6.5) is based on the lahar stratigraphy and mapped extent of laharic surfaces in the catchment as described above. Each lahar hazard zone was derived from a mapped laharic surface of known or inferred age. Each assignment is based on the age of the youngest lahar to cross the surface.

136 6.10.1 1 in >15 000 year zone

This zone includes the most extensive surfaces in the Tongariro catchment. The laharic surfaces included within this zone comprise lahar deposits older than the Rerewhakaaitu Tephra (14.7 ka B.P.). Almost all of the eastern ring plain of Tongariro volcano and large parts of the northeastern Ruapehu ring plain comprise laharic surfaces of this age. In the southern part of the mapped area, near the Upper Waikato Stream, the zone includes lahar surfaces which are pre-Kawakawa Tephra age. A small patch of a pre-Kawakawa Tephra surface is also preserved immediately south of Turangi township (Fig. 6.5). On the Ruapehu ring plain the lahar deposits within this hazard zone belong to the Te Heuheu Formation (Donoghue, 1991 ). On Tongariro volcano these lahar deposits were previously mapped as the Rangipo Lahars (Grindley, 1960). This extensive zone and the large number of lahars that comprise these surfaces (Table 6.2 and Cronin and Neall, in press) indicate that lahars were a common occurrence across large areas of both ring plains during and following the Last Glacial Maximum. At risk within this zone is the village of Rangipo and part of the Rangipo Prison, State Highway 1 and other secondary roads. In addition, high voltage transmission lines and the Poutu Canal cross this zone.

6.1 0.2 1 in 12 000 year zone

This zone covers small areas alongside the Oturere and Makahikatoa Streams on the Tongariro ring plain and lower slopes. lt represents a lahar path following deposition of the Rotorua Tephra but prior to the eruption of Waiohau Tephra (1 1 .85 ka B.P.). The zone is crossed by State Highway 1 and high voltage transmission lines, but mostly includes unpopulated land covered by native forest.

6.1 0.3 1 in 10 000 year zone

This zone represents a lahar surface recognised by its oldest marker coverbed being either the Pourahu, Pahoka, or Poronui Tephra, all of which are between the ages of 9.8- 10 ka B.P. This surfa ce covers a sector of the northeastern Ruapehu ring plain, and comprises deposits of several post-glacial lahars which flowed down the Te Piripiri, Mangatoetoenui, and Waihohonu Streams (Table 6.2). The surface pinches out in the central reaches of the Tongariro River where lahars of this age were probably confined to an incised channel. Immediately upstream of the Rangipo prison the lahar surface again spreads out to cover a large area between Rangipo and Turangi. Much of Turangi township is built on this surface as well as the Rangipo Prison and a few other outlying houses. Also at risk on the surface are numerous roads, telephone lines, high voltage transmission lines, as well as large areas of pastoral land and livestock.

137 175° 40' 50' Lake Ta upo Lahar hazard Zones Mangatoetoenui . 1 In 35 ye a rs Stream 1 in 1 00 years 1 000 years 2 000 years 39o 5 000 years 39° 00' 1 0 000 years 00' 12 000 years >15 000 years

N Pihanga

A •

05'

Figure 6.4 Lahar hazard map of the Tongariro River. Locations of sections in Figs. 6.2 and 6.3 are labelled.

15' I I --- Major roads I --- Minor roads I Power development tunnels I I 0 2 3 4 5 km I 45' 50'

till �------

6.1 0.4 1 in 5 000 year zone

Two small areas of this hazard zone are delineated, along Mangatoetoenui Stream (Fig. 6.4) and a small area south of Turangi township (Fig. 6.5). These surfaces were last inundated by a lahar between the time of the eruptions of the Hinemaiaia and Motutere Tephras (Table 6.2, c. 5.2-5.3 ka B.P.). Both of these areas are unpopulated and are not used for agricultural production.

6.1 0.5 1 in 2 000 year zone

Surfaces within this zone are extensive only in the lower reaches of the Tongariro River (Fig. 6.5). The last lahar to inundate this surface was immediately prior to the Taupo Tephra eruption. Upstream of Turangi these surfaces are confined close to the present river channel but fan outwards north of the town. Parts of Turangi, and Tokaanu and many outlying houses are built on this surface. An airstrip, the Turangi sewage treatment area, the Tongariro trout hatchery and many roads (including State Highway 1) are also within this zone. Large portions of the zone are also in pastoral production.

6.1 0.6 1 in 1 000 year zone

This zone includes areas which are interpreted to have been crossed by a lahar since the eruption of the Taupo Tephra (1.85 ka B.P.). Only a few localities within this zone preserve deposits of a post-Taupo Tephra lahar because many of the deposits have been reworked by subsequent fluvial action. The area covered by this zone extends from Turangi to the delta in Lake Taupo (Fig. 6.5). In the Tongariro delta region, where exposure is poor, the extent of the zone has been delimited by consideration of the soil map of this region (Water and Soils Division, Ministry of Works and Development, 1979). The mapped extent of the 1 in 1 000 year zone corresponds to "Recent soils from alluvium", whilst "Organic soils" beyond the limit of these lahars, occur in the surrounding swampy area. Parts of Turangi are included within the zone as well as a few outlying houses and the State Highway 1 Bridge across the river.

6.10.7 1 in 100 year zone

This zone comprises the channel of the Tongariro River into which at least one lahar flowed during the 1995 eruption sequence of Ruapehu (Cronin et al., in press(b)). Europeans did not settle in the Turangi region until the early 1900's (Ciarke and Smith, 1986), so that continuous records were not available until this time. Although the 1995 lahar did not pass beyond the Rangipo Dam in the Tongariro River, a 1 in 100 year hazard zone is still applied to the remainder of the river channel because this would probably be the case without human intervention.

139 50' Lake Ta upo

To kaanu

N A To kaanu Power Station 39° 00'

To ngariro River 1 in 100 years channel 1 in 1 000 years 1 in 2 000 years 1 in 5 000 years 1 in 1 0 000 years 1 in >15 000 years

03' .. Pihanga

--- Major roads --- Minor roads

0 2km

Figure 6.5 Enlargement of lahar hazard map for Turangi and its surrounds. This area contains the greatest population and most of the youngest laharic surfaces in the Tongariro catchment. Location of sections in Figs. 6.2 and 6.3 are labelled.

140 6.10.8 1 in 35 year zone

This zone includes the channel of the Mangatoetoenui Stream which was affected by at least two lahars in the 1995 eruption sequence of Ruapehu (Cronin et al., in press(b )), and also by lahars in the 1895 and 1975 Ruapehu eruptions (Alien, 1902; Nairn et al., 1979). Observations of Ruapehu volcano have been made for ea. 140 years (Gregg, 1960), and four lahars recorded during this time equates to a 1 in 35 year recurrence interval.

6.11 Discussion

The areas of greatest lahar hazard in the catchment are obviously the lowest lying areas closest to the Tongariro River and Mangatoetoenui Stream. The area of greatest risk in the catchment is where the highest lahar hazard and greatest vulnerability of people and property are combined. Although the entire town of Turangi is built on laharic surfaces, much of it is on a surface which has not been inundated by lahars since 10 ka B.P. However, parts of Turangi and many outlying houses are on surfaces which have been covered by much younger lahars. Thus, the area of greatest human risk in this catchment is where houses are built within the 1 in 1 000 year hazard zone in part of Turangi. One of the greatest property risks in the area has already been identified by lahar impacts during the 1995 Ruapehu eruption sequence. The Rangipo Dam became substantially infilled with sediment from lahars sourced within the Mangatoetoenui catchment. This dam retains water to be used for power generation in the underground Rangipo Power Station. Sandy and silty sediment within water piped to the power station led to severe wear of the turbine blades, necessitating replacement of one of them (ECNZ pers. comm., 1995). The State Highway 1 bridge crossing the Mangatoetoenui Stream on the Desert Road is probably the bridge at greatest risk from lahars in the Tongariro catchment. The source of most lahars in the period 14.7-10 ka B.P. in the Tongariro catchment has been Ruapehu volcano, particularly via the Mangatoetoenui and Te Piripiri Streams. All Holocene lahars have also come from Ruapehu and most of these were sourced within Mangatoetoenui Stream. Lahars in the Tongariro catchment have been attributed to a variety of generation mechanisms, including flank collapse, eruptions, glacier collapse and storms. The most readily identifiable are those related to tephra eruptions of either Tongariro or Ruapehu (Cronin and Neall, in press). Almost all of the Holocene lahars have been linked to large scale tephra eruptions from either volcano. The 1995 lahars followed this pattern, but were of a smaller scale than most Holocene lahars. Unlike the Whangaehu River, the Tongariro River tributaries do not directly drain Crater Lake, nor are they as close to the Lake as the Mangaturuturu, Whakapapaiti and Whakapapanui catchments. Thus, lahars generated by water expelled during small scale eruptions through the Lake have not occurred with the same frequency in the Tongariro

141 tributaries as they have in the other catchments. Two lahars derived from surges from Crater Lake were produced in the Mangatoetoenui Stream during eruptions in 1895 and 1975. In contrast, 19 lahars derived from Crater Lake are recorded in the Whangaehu Catchment prior to 1995 (Hodgson, 1993), and a further 26 occurred during the 1995 Ruapehu eruptive sequence (Cronin et al., in press(b)). During 1995, events in the Mangatoetoenui Stream indicate another mechanism for creating lahars that enter the Tongariro River during eruptions of Ruapehu. These lahars were caused when large amounts of tephra fell onto Mangatoetoenui Glacier, following several eruptions throughout September and early October 1995. The largest of these eruptions on October 11-12 contributed the greatest volume of tephra. Heavy rains in the following weeks caused collapse, rilling and remobilisation of saturated tephra to form lahars (Cronin et al. , in press(b)). A further potential lahar generating mechanism in this catchment is the eruption of hot pyroclastic flows onto the Mangatoetoenui Glacier, which could rapidly entrain and melt snow and ice. Lahars caused by these mechanisms occurred on Nevado del Ruiz, where relatively small scale eruptions led to huge, devastating lahars (Pierson et al., 1990). Evidence of only one pyroclastic flow has been recorded on Ruapehu at c. 10 ka B.P. (Donoghue et al., 1995) but most notably a lahar associated with this event is recorded in the Mangatoetoenui catchment (Cronin and Neall, in press). Another possible way for lahars to enter the Tongariro River is via overflow of a large lahar from the Whangaehu River into Upper Waikato Stream. Deposits recording such an event have not been described in this study, but are described by Donoghue (1991). During the 1995 eruption sequence of Ruapehu this situation almost eventuated. Several lahars between September 18 and 25 gradually built up the base level of the Whangaehu channel on the upper ring plain by several metres until on September 25 the largest lahars flooded into small tributaries which flow northwards before rejoining the Whangaehu River farther downstream. These northeastward-flooding lahars travelled to within 300 m of tributaries of Upper Waikato Stream. If diversion of the main Whangaehu River channel into this route were to occur, lahars would probably overflow into Upper Waikato Stream. Incision of the Whangaehu River through the aggraded lahar deposits over the following few months has temporarily reduced this risk. If the northeastward flowing channels become further incised during heavy rainstorms, headward erosion could lead to capture of the main Whangaehu River channel. A northeastward course of the River would then lead to a new route within 300 m of Upper Waikato Stream. Hence the potential for riverbank erosion and overflow northwards could lead to future lahars being directed into the Tongariro River.

6.11.1 Mitigation of hazards

Three areas are identified where hazards of lahars in the Tongariro River can be mitigated.

142 ------

1) Warning of lahars. Given adequate warning of a lahar, areas at risk can be evacuated and loss of life minimised. Currently there is no lahar warning device installed on the Tongariro River or any of its tributaries. Although monitoring of flow levels is carried out by ECNZ for managing the Tongariro Power Development, a purpose-built lahar warning gauge would provide the best and most timely warning. 2) Containment of small lahars. Lahars of a certain volume could be potentially retained behind the Rangipo Dam, particularly if the dam were emptied in preparation for such a lahar. The same strategy was advised and utilised in the Swift Reservoir to contain a lahar from Mt. St. Helens in 1980 (Crandell and Mullineaux, 1978; Schuster, 1981). Containment of a lahar in the Rangipo Dam would create problems for the power development but could potentially reduce downstream impacts. However, if the lahar were larger than the dam capacity, the facility could be destroyed in attempting to contain it. 3) Planning of future expansion of Turangi. Parts of Turangi have expanded onto surfaces within the 1 in 2000 and 1 in 1 000 year lahar hazard zones across the River from the main area of town. In planning future housing expansion of the town, building within the 1 in 10 000 year zone should be encouraged over areas of greater hazard.

6.12 Conclusions

The greatest number and thickness of lahar deposits in the exposed record of the Tongariro catchment are from lahars occurring between 14.7 and 9.8 ka B.P. Consequently the preserved laharic surfaces of greatest area are also of this age. Deposits of Holocene lahars are confined to restricted surfaces, closer to the present river channel. These deposits are less well preserved than the older units because they were probably confined to a more deeply incised channel than during late glacial and early post-glacial times. The town of Turangi and many outlying houses are built on surfaces that are laharic in origin. However, lahars have not inundated a large proportion of the town since c. 10 ka years B.P. Lower-lying parts of Turangi, closer to the Tongariro River and other outlying houses are built on surfaces which have been inundated by much younger lahars. The area identified with the greatest human risk are the lowest-lying parts of Turangi, which were last inundated by a lahar less than 1850 years B.P. Mitigation of this risk could be provided by a lahar warning system and careful planning of future urban development in Turangi. The Mangatoetoenui Stream tributary is identified as the most important lahar conduit for the Tongariro catchment during the Holocene. lt has also been the only lahar conduit in this catchment in historic times. Consequently, infrastructure at greatest risk of damage from lahars includes the State Highway 1 bridge crossing Mangatoetoenui Stream, the Rangipo Dam, the Rangipo Power Station that it supplies, and the Poutu Canal intake structure.

143 -- -- � ------�------�- -

6.13 Acknowledgements

SJC gratefully acknowledges funding from the New Zealand Vice-Chancellor's Committee, Massey University Graduate Research Fund, the Helen E. Akers Scholarship Fund and the Tongariro Natural History Society. We thank J.A. Lecointre for his comments on an earlier version of the manuscript.

References

Alien, G. F., 1902. Supplementary edition of Willis' guide book of new routes for tourists. Willis, Wanganui: 240 pp. Beverage, J.P., Culbertson, J.K., 1964. Hyperconcentrations of suspended sediment. J. Hydraul. Div., American Soc. of Civil Engineers, 90: 117-126. Clarke, C., Smith, M., 1986. Turangi town centre, where to now? Taupo County Council Special Report, 15. Crandell, D.R., Mullineaux, D.R., 1978. Potential hazards from future eruptions of Mount St. Helens volcano, Washington. U.S.G.S. Bull., 1383-C: 26 pp. Cronin, S.J., Neall, V.E., in press. A late Quaternary stratigraphic framework for the northeastern Ruapehu and eastern Tongariro ring plains, New Zealand. N. Z. J. Geol. and Geophys., 40. Cronin, S.J., Neall, V.E., Stewart, R.B., Palmer, A.S., 1996. A multiple parameter approach to andesitic tephra correlation, Ruapehu volcano, New Zealand. J. Volcano!. and Geotherm. Res., 72: 199-215. Cronin, S.J., Neall, V.E., Lecointre, J.A., Palmer, A.S., in press(b). Changes in Whangaehu River lahar characteristics during the 1995 eruption sequence, Ruapehu volcano, New Zealand. J. Volcano!. and Geotherm. Res. Cronin, S.J., Neall, V.E., Palmer, A.S., Stewart, R.B., in press(a). Methods of identifying late Quaternary rhyolitic tephras on the ring plains of Ruapehu and Tongariro volcanoes, New Zealand. N. Z. J. Geol. and Geophys, 40. Department of Statistics, 1992. 1991 Census of Population and Dwellings, Waikato/Bay of Plenty Regional Report. Department of Statistics, Wellington, N.Z. Donoghue, S.L., 1991. Late Quaternary volcanic stratigraphy of the south-eastern sector of Mount Ruapehu ring plain, New Zealand. Unpub. PhD thesis, Massey University, Palmerston North, New Zealand. Donoghue, S.L., Neall, V.E., Palmer, A.S., 1995. Stratigraphy and chronology of late Quaternary andesitic tephra deposits, Tongariro Volcanic Centre, New Zealand. J. Roy. Soc. of N. Z., 25: 115-206. Froggatt, P.C., Lowe, D.J., 1990. A review of late Quaternary silicic and some other tephra formations from New Zealand: their stratigraphy, nomenclature, distribution, volume, and age. N. Z. J. Geol. and Geophys, 33: 89-109. Gregg, D.R., 1960. The geology of Tongariro Subdivision. N.Z. Geol. Survey Bull., 40: 151 pp.

144 Grindley, G.W., 1960. Geological map of New Zealand, 1st ed., 1: 250 000, Sheet 8, Taupo. Dept. of Scientific and Industrial Res., Wellington. Hackett, W.R., Houghton, B.F., 1989. A facies model for a Quaternary andesitic composite volcano: Ruapehu, New Zealand. Bull. Volcano!., 51: 51-68. Hodgson, K.A., 1993. Late Quaternary lahars from Mt. Ruapehu in the Whangaehu River valley, North Island, New Zealand. Unpub. PhD thesis, Massey University, Palmerston North, New Zealand. Mathews, W.H., 1967. A contribution to the geology of the Mount Tongariro massif, North Island, New Zealand. N. Z. J. Geol. and Geophys., 10: 1027-1038. Nairn, I. A., Self, S., 1978. Explosive eruptions and pyroclastic avalanches from Ngauruhoe in February 1975. J. Volcano!. and Geotherm. Res., 3: 39-60. Nairn, I.A., Wood, C.P, Hewson C.A.Y., 1979. Phreatic eruptions of Ruapehu: April 1975. N. Z. J. Geology and Geophys., 22: 155-173. Palmer, B.A., Purves, A.M., Donoghue, S.L., 1993. Controls on the accumulation of a volcaniclastic fan, Ruapehu composite volcano, New Zealand. Bull. Volcano!., 55: 176-189. Philippine Institute of Volcanology and Seismology, 1992. Pinatubo: Lava dome growth; pyroclastic flow deposits spawn destructive lahars and secondary explosions. Bull. Global Vole. Network 17(9): 8-10. Pierson, T.C., Costa, J.E., 1987. A rheologic classification of subaerial sediment-water flows. G.S.A., Reviews in Engineering Geol. VII: 1-12. Pierson, T.C., Janda, R.J., Thouret, J.-C., Borrero, C.A., 1990. Perturbation and melting of snow and ice by the 13 November 1985 eruption of Nevado del Ruiz, Colombia, and consequent mobilisation, flow and deposition of lahars. J. Volcano!. and Geotherm. Res., 41: 17-66. Pierson, T.C., Janda, R.J., Umbal, J.V., Daag, A.S., 1992. Immediate and long-term hazards from lahars and excess sedimentation in rivers draining Mt. Pinatubo, Philippines. U.S.G.S. Water Resources Investigations Rep., 92-4039: 182-203. Pierson, T.C., Scott, K.M., 1985. Downstream dilution of a lahar: transition from debris flow to hyperconcentrated streamflow. Water Resources Res., 21: 151 1-1524. Schuster, R.L., 1981. Effects of the eruptions on civil works and operations in the Pacific Northwest. In: Lipman, P.W.; Mullineaux D.R. (Eds.) The 1980 eruptions of Mount St. Helens, Washington. U.S.G.S. Prof. Pap. 1250: 701-718. Smith, G.A. 1986: Coarse-grained nonmarine volcaniclastic sediment: Terminology and deposition process. G.S.A. Bull., 97: 1-10. Smith, G.A., Fritz, W.J., 1989. Volcanic influences on terrestrial sedimentation. Geology 17: 375-376. Stilwell, W.F., Hopkins, H.J., Appleton, W., 1954. Tangiwai railway disaster. Report of the Board of Inquiry. Government Printer, Wellington: 31 pp. Topping, W.W., 1973. Tephrostratigraphy and chronology of late Quaternary eruptives from the Tongariro Volcanic Centre, New Zealand. N. Z. J. Geol. and Geophys., 16: 397-423.

145 Topping, W.W., Kohn B.P., 1973. Rhyolitic tephra marker beds in the Tongariro area, North Island, New Zealand. N. Z. J. Geol. and Geophys., 16: 375-395. Voight, B., 1990. The 1985 Nevado del Ruiz volcano catastrophe: anatomy and retrospection. J. Volcano!. and Geotherm. Res., 44: 349-386. Water and Soils Division, Ministry of Works and Development, 1979. N. Z. Land Resource Inventory Worksheets, N102 Tokaanu and N112 Ngauruhoe and Bay of Plenty - Volcanic Region Landuse Capability Extended Legend (12pp}. National Water and Soil Conservation Organisation, Ministry of Works and Development, Wellington, N.Z. Wilson, C.J.N., Switzur, R.V., Ward, A.P., 1988. A new 14C age for the Oruanui (Wairakei) eruption, New Zealand. Geol. Mag., 125: 297-300. Wilson, C.J.N., Houghton, B.F., Lanphere, M.A., Weaver, S.D., 1992. A new radiometric age estimate for the Rotoehu Ash from Mayor Island volcano, New Zealand. N. Z. J. Geol. and Geophys., 35: 371 -374. Wilson, C.J.N., 1993. Stratigraphy, chronology, styles and dynamics of late Quaternary eruptions from Taupo Volcano, New Zealand. Phil. Trans. Roy. Soc. of London A 343: 205-306.

146 CHAPTER 7: GEOLOGY AND LANDSCAPE DEVELOPMENT OF THE NORTHEASTERN TONGARIRO VOLCANIC CENTRE

7.1 Introduction

In the three previous chapters, rhyolitic and andesitic tephrostratigraphy (developed in chapters 2 and 3) were used to establish a climate/soil development history and a history of lahars and landscape development by lahars in the study area. This chapter combines the results of the previous chapters and adds further information on other landscape and geological events on the ring plains to provide an overall synthesis of their geology and development. This chapter comprises three parts: (7.2) a geological map of the northeastern Tongariro Volcanic Centre with accompanying notes, (7.3) a published paper entitled "Geological history of the northeastern ring plain of Ruapehu volcano, New Zealand", and (7.4) a summary geological synthesis of the entire study area. The paper: Geological history of the northeastern ring plain of Ruapehu volcano, New Zealand by: Shane J. Cronin, V.E. Neall and A.S. Palmer, is published within Quaternary International Vol. 34-36. The contributions of each author to the study were as follows.

S. J. Cronin: Principal Investigator Carried out all: Field mapping, description and sampling Mineralogy analyses All other laboratory studies Preparation and writing of manuscript

V. E. Neall and A. S. Palmer: Advisors Aided the study by: discussion of results and methodology editing and discussion of the manuscript

7.2 Surficial geologic map of the northeastern Tongariro Volcanic Centre

The surficial geologic map of the northeastern Tongariro Volcanic Centre is presented in Fig. 7 .1. Figure 7.2 demonstrates the stratigraphy of the mapping units and their relationship to tephra marker beds and previously mapped formations in the surrounding area.

7.2.1 Methods

The volcanic and volcaniclastic deposits and sediments within the area were mapped using field sheets of 1 :25 000 scale. Field-section descriptions were supplemented with vertical aerial photographic interpretation to delineate surfaces. Stereo-pairs of 1981 black and white vertical aerial photographs at 1:25 000 scale were used.

147 MAP OF THE SURFICIAL GEOLOGY OF A PORTION OF THE NORTHEASTERN TONGARIRO VOLCANIC CENTRE 600 m'-1 To kaan Power Station 39° 39° 00' 00'

Legend

Laharic surfaces

< 1.85 ka. > 1.85 ka > 5.3 ka > 9. 7 ka, Ruapehu > 12 ka, To ngariro 1-- !....-i 14.7 ka, Ruapehu ----' >

L...___ > 14.7 ka, To ngariro and _.J Ruapehu

Lavas

< 14.7 ka, To ngariro 1-"--"-"---'--'--'i > 14.7 ka, Ruapehu 1--::-:::----1 > 14.7 ka, To ngariro 1-- -1 ='---- > 22.6 ka, To ngariro .______.J Region Mangahouhounui not mapped Stream Moraines

- > 1 0 ka moraines

Composite surfaces

Rangipo > 14.7 ka Tongariro lahar surface, covered Power by < 1.85 ka alluvium. Station

> 14.7 ka lahar surface veneering > 22.6 ka lava, Tongariro.

> 14.7 ka Ruapehu lahar surface, reworked and partially covered by < 10 ka alluvium.

Symbols

0 Covered by up to 30 m 0 or Taupe ignimbrite

Major roads 900 m Minor roads Wa ipakihi River Power development 15' tunnels To pographic contours I 1 00 m intervals Upper Wa ikato State highways Stream

0 1 2 3 4 5km

I Te phras Lahars La vas Moraines Age ka B.P. Ruapehu This study To ngariro This study This study (Fig. 7.1) (Donoghue, 1991) Fig. 7.1 (Grindley, 1960) Fig. 7.1 Ruapehu To ngariro 0 I Onetapu ? ? Ta upo Te phra I =L2

Waimahia Te phra (Un' S) 1 Hinemaiaia Te phra (Unit R) I 4 Manutah Hinemaiaia Te phra (Unit 11-=:::o Motutere Te phra (Unit H) 6 Motutere Te phra (Unit G) � I Whakapapa Formation 8 Poronui Te phra (Unit Ta ngatu Formation Mangamate Te phra C Karapiti Te phra (Unit �====�==�====��======� 1 0 Pahoka Te phra To ngariro Pourahu Member Andesite Wa iohau Te phra -----+-----+----- �---=i- 12 Rotorua Te phra Rotoaira Te phra ------+---- -+-----1 14 Rerewhakaaitu Te phra f I 16 Mangawhero Formation 18 Te Heuheu Formation

22 Kawakawa Te phra I l Rangipo Lahars (Penultimate 24 Glacial)

Fig 7.2 Legend of geologic units mapped in Fig. 7.1, showing their stratigraphic relationships to previously mapped formations and tephra marker beds. The scale of the map presented in Fig. 7.1 is 1:100 000. The base map was prepared by scanning and digitising contours, rivers, roads and other features from NZMS 260 series topographic maps (1 :50 000 scale}, T19-Edition 2 (DOSLI, 1994) and T20-Edition 1 (DOSLI, 1982). The study area was mostly limited to the ring plain and lower slopes of the Ruapehu and Tongariro volcanoes but also included the lower Tongariro River valley. Areas were delineated and mapped according to their age (determined by cover-bed stratigraphy), their lithology, and the volcano from which they were derived (Tongariro or Ruapehu}.

7.2.2 Laharic surfaces

Seven laharic surfaces are mapped in the studied area. Each surface is delimited by the deposits of the youngest lahar preserved on it. The ages of these lahars are interpreted from tephra marker beds which under- and overlie them (Fig. 7.2). The stratigraphy of these surfaces was presented in Chapters 5 and 6.

7 .2.2A Ruapehu-derived lahar surfaces The youngest lahar surface (<1 .85 ka B.P.) has been inundated by at least one lahar since the eruption of the Taupo Tephra. In Chapter 6 it was postulated that late Holocene lahars in the Tongariro catchment were derived from Ruapehu volcano. Thus lahar deposits forming this surface are correlated to the Onetapu Formation (after Donoghue, 1991; Hodgson, 1993). The > 1.85 ka surface, if it were derived from the same source, is between Onetapu Formation and Manutahi Formation in age (Fig. 7.2) and comprises a newly recognised Ruapehu lahar surface. Ruapehu-sourced lahars mapped by Donoghue (1991) as the Tangatu Formation, are here mapped as two separate surfaces. The youngest and smaller surface comprises a lahar deposit in the order of 5.2-5.3 ka (R1 , Chapter 5). This surface is preserved along the Mangatoetoenui Stream and a small remnant is preserved immediately south of Turangi. The other Tangatu Formation surface mapped here comprises multiple lahars which are > 9.7 ka in age. This surface occurs on the northeastern Ruapehu ring plain between the Te Piripiri and Ohinepango Streams and is also mapped over large areas from Turangi south to Rangipo. Large areas of Te Heuheu Formation occur on the northeastern Ruapehu ring plain between Te Piripiri and Upper Waikato Streams and are mapped as the > 14.7 ka (Ruapehu) surfa ce. A portion of this surface has been eroded and reworked by the Holocene/recent activity of streams crossing the area and so is mapped as a composite surface. North of the Waihohonu Stream, throughout the area where the Tongariro River cuts through the Tongariro volcano ring plain, Te Heuheu Formation lahar deposits are also interbedded with contemporaneous lahar deposits derived from Tongariro volcano. In this area it is impossible to differentiate the lahar deposits from the two volcanoes or a boundary between them. Consequently, this surface is mapped as containing lahars from both volcanoes, but is probably dominated by Tongariro-sourced lahars.

150 7.2.28 Tongariro-derived /ahar surfaces The lahars of the Tongariro ring plain were mapped by Grindley (1960) as the Rangipo Lahars and were thought to be of Penultimate Glacial age. Here the eastern Tongariro ring plain is mapped mostly as Last Glacial-age lahar deposits with smaller areas of early post-glacial lahars. The youngest lahar surfaces derived from Tongariro volcano comprise lahar deposits which are > 12 ka in age (Fig. 7.2). These surfaces are mapped beside the upper portion of the Waihohonu Stream and along the Oturere and Makahikatoa Streams. These lahar deposits interdigitate with and are covered by Tangatu Formation lahar deposits (derived from Ruapehu volcano) in the lower Waihohonu Stream and along the Tongariro River. The remaining eastern Tongariro ring plain mostly comprises surfaces composed of lahar deposits > 14.7 ka in age. This is mapped in two areas. The first area is mapped as a simple lahar surface comprising dominantly Tongariro volcano-derived lahar deposits > 14.7 ka in age. This surface is mapped on the Tongariro ring plain sector between the Mangamate Stream and Lake Rotoaira. This surface probably also includes subordinate lahar deposits derived from Ruapehu volcano, particularly in areas close to the Tongariro River. The second area (between the Waihohonu and Mangamate Streams) is mapped as a composite surface. This surface comprises lahar deposits >14.7 ka in age which veneer Tongariro-sourced lava flows which are >22.6 ka B.P. in age. In the present stream valleys the lahar deposit cover can be up to tens of metres thick, whilst on some ridge tops and other isolated areas the lahar veneer is absent. The lahar deposits are mostly derived from Tongariro volcano, but probably also include lahar deposits derived from Ruapehu volcano, particularly at the southern part of the mapped area and near the Tongariro River.

7.2.3 Lavas

Lava flows derived from both volcanoes were mapped on the ring plains and were dated by their tephra coverbeds. The petrology of some the lavas was also described in order to characterise them and to correlate exposures.

7 .2.3A Ruapehu-derived /avas

On the northeastern Ruapehu ring plain several lava flows are mapped. All of these flows have similar coverbed sequences and their oldest coverbed is the Rerewhakaaitu Tephra (Fig. 7.2). These lava flows also have very similar lithologies. They all have a porphyritic aphanitic texture and contain abundant glomerocrysts. They are all two-pyroxene andesites, with a phenocryst assemblage dominated by large, lath-shaped plagioclase crystals {Table 7.1 ). Plagioclase phenocrysts display strong oscillatory zoning. Orthopyroxene is the dominant ferromagnesian phenocryst with around twice the modal % of clinopyroxene.

151 Table 7.1 Mineralogy and locations of lava sampled from the ring plains of northeastern Tongariro Volcanic Centre. Sample NZMS 260 Mapping Unit Phenoc!Yst modal % Location reference PL CPX OPX TM OL

93.14 T20/442146 > 14.7 ka Ruapehu 60 17 18 5 - Flow north of Mangatoetoenui Stream 93.165 T20/448159 " 46 19 35 <1 " 93.23 T20/446127 62 11 26 1 - Flow north of Te Piripiri Stream 93.106 T20/463122 " 57 13 29 1 - Flow along Wharepu Stream 93.107 T20/435130 " 53 15 32 <1 - Flow south of Te Piripiri Stream

95.5 T19/420243 < 14.7 ka Tongariro 32 24 30 10 3 Upper Oturere valley " 95.8 T19/423339 40 17 37 6 - Flow Southwest of Lake Rotoaira

93.19 T20/464174 > 22.6 ka Tongariro 53 18 26 3 - Flow in mid-Waihohonu Stream beside State Highway 1

93.16 T20/492185 > 14.7 ka laharic surface veneering 62 19 19 <1 - Flow within lower Waihohonu Stream >22.6 ka lava, Tongariro 94.8 T20/506185 " 54 12 31 1 2 Within Tongariro River channel, downstream of Rangipo Dam " 94.31 T20/499178 58 9 32 1 - At Rangipo Dam 94.33 T19/528227 " 50 13 35 2 - Tongariro River, downstream of Tree Trunk Gorge 94.36 T19/520221 " 34 16 49 1 - Tongariro River, upstream of Tree Trunk Gorge 94.55 T19/506238 " 63 32 4 1 - Flow in lower Mangatawai Stream 94.75 T19/535246 " 41 10 44 1 4 Tongariro River, Pillars of Hercules 95.16 T19/550331 " 44 19 33 1 3 Tongariro River, on Rangipo Prison Farm

95.10 T19/448326 > 22.6 Tongariro 11 32 3 - South of Lake Rotoaira 95.12 T19/466298 " 54 8 29 9 - North of Mangahouhounui Stream 54 94.80 T19/536376 Unmapped Pihanga flow <1 14 20 1 65 Tongariro River at Poutu Pool

PI = Plagioclase, CPX = Clinopyroxene, OPX = Orthopyroxene, TM = Titanomagnetite, OL = Olivine.

152 Titanomagnetite occurs in association with pyroxene phenocrysts and within glomerocrysts. Glomerocrysts are made up of 20 to 1 00+ crystals of all four phenocryst phases. Two types of glomerocrysts were observed, those comprising few (20-30) large crystals, and others comprising larger numbers of smaller crystals which were dominantly plagioclase. The groundmass is dominated by (80-95% by volume) small laths of plagioclase with minor orthopyroxene and clinopyroxene crystals. Based on their mineralogy, these lavas appear to belong to the Type 1 class (plagioclase-pyroxene phyric lavas) of Graham and Hackett (1987). The minimum age of these lavas (determined from their oldest coverbed, Fig. 7.2) indicate that these flows correlate to the Mangawhero Formation of Hackett (1985).

7.2.38 Tongariro-derived lavas

The Tongariro massif was previously mapped as a single unit, Tongariro Andesite, which spanned the Quaternary in age (Grindley, 1960). In this study, three Tongariro-derived lava surfaces are mapped in addition to a composite surface comprising lavas and lahar deposits. The oldest lava surfaces are those overlain by the Kawakawa Tephra, indicating a minimum age of 22.6 ka B.P. The lava flow beside the Waihohonu Stream has the Okaia Tephra below it, providing an age bracket of 22.6 - c. 23 ka (Section 5.1 0). The other lava surface mapped with the same minimum age (northeastern Tongariro volcano) has K-Ar ages ranging between 105 and 130 ka for samples taken higher on the cone (Hobden et al. , 1996). If one believes the radiometric ages, then this indicates that the northeastern Tongariro lava surface has an unconformable coverbed sequence. Samples from both areas have a very similar mineralogy to one another and to the Ruapehu lavas described in the last section (Table 7.1 ). However, these samples have a finer-grained, plagioclase-dominated groundmass, and glomerocrysts comprising fewer crystals (1 0-20) than the Ruapehu flows described above. These lavas also fit within the Type 1 classification of Graham and Hackett (1987). Lavas comprising the surface which is mapped as the composite of >14.7 ka lahars veneering >22.6 ka lava, also derive their minimum age from being overlain by the Kawakawa Tephra. However, given that this sector of the Tongariro cone is K-Ar dated at between 65 and 130 ka (Hobden et al., 1996) these lavas are probably of similar or greater age. Samples from some of these flows have a pyroxene-dominant mineralogy rather than plagioclase-dominant. Three of the samples also contain a small percentage of large olivine phenocrysts. All samples have a very fine-grained groundmass comprised of laths of plagioclase and interstitial brown glass. These mineralogical features indicate that some of these lavas were more basic than those described above. These lavas include both Type 1 (plagioclase-pyroxene phyric lavas) and Type 3 (pyroxene-phyric lavas) classifications of Graham and Hackett (1987) for Tongariro Volcanic Centre lavas. The > 14.7 ka {Tongariro) lavas mapped are overlain by the Rerewhakaaitu Tephra, indicating their minimum age (Fig. 7.2). However, samples from these flows

153 higher on the cone give K-Ar ages of 65-130 ka (Hobden et al., 1996), indicating the coverbed sequences are potentially unconformable. No samples of these lavas were collected. The disparity between the K-Ar ages of Hobden et al. (1996) and the coverbed sequences on the Tongariro lava surfaces, implies that these surfaces were exposed to intense physical weathering during the last stadia! of the Last Glacial. Two lava flows are mapped as < 14.7 ka. The flow southwest of Lake Rotoaira is immediately overlain by the Waiohau Tephra but the .Rerewhakaaitu Tephra (present on the adjacent lava surface) does not occur. This provides an age range for this lava flow of between 14.7-1 1.9 ka B.P. The other flow mapped as < 14.7 ka occurs within the Oturere Valley. No coverbeds were found overlying this lava flow, which may indicate it is very young. Both of these two flows have a Type 3, pyroxene-dominant mineralogy, and the Oturere flow also contains olivine (Table 7.1 ).

7.2.4 Moraines

Parts of two sets of paired moraines were mapped in the study area based on their morphology. The moraines beside the Waihohonu Stream have been previously described by Mathews (1967) and those flanking the Mangatoetoenui Stream by McArthur and Shepherd (1990). Mathews (1967) estimated that the Waihohonu moraines were around 15 ka in age by comparison to the youngest moraines in the South Island. McArthur and Shepherd (1990) considered the Mangatoetoenui moraines to have been formed in multiple glacier advances and represented the last two stadials of the Last Glaciation. In this study, tephra coverbeds on these two sets of moraines provide minimum ages for their stabilisation and the cessation of their construction (Fig. 7.2). The Waiohau Tephra occurs on the Waihohonu moraines providing a minimum age of stabilisation of 11.9 ka B.P. The oldest tephra on the Mangatoetoenui moraines is the Pourahu Member of the Bullet Formation indicating a minimum age of ea. 10 ka.

7.3 GEOLOGICAL HISTORY OF THE NORTH-EASTERN RING PLAIN OF RUAPEHU VOLCANO, NEW ZEALAND

Shane J. Cronin, Vincent E. Neall and Alan S. Palmer Department of Soil Science, Massey University, Private Bag 11 222, Palmerston North, NewZealand

1996, Quaternary International 34-36: 21-28.

Abstract Continuous exposures along the Upper Waikato Stream provide new insights into the north-eastern ring plain of Ruapehu volcano, extending the known stratigraphy beyond 22.5 ka. Time control in the sequence is provided by five rhyolitic tephra units, erupted from central North Island volcanoes, comprising Kawakawa, Okaia,

154 Omataroa, Hauparu, and Rotoehu tephras. The sequence is dominated volumetrically by diamictons and fluvial deposits resulting both from volcanic events and periods of instability on the flanks of Ruapehu. Within the sequence are >60 individual andesitic lapilli units, derived primarily from Ruapehu volcano via mostly sub-plinian eruption mechanisms. An average eruption rate of more than one lapilli eruption per 1000 years is estimated for the c. 60 ka record. The style of deposition on the ring plain changes over time and appears to reflect climate change over the Last Glacial period. In periods of severe climatic conditions during marine &180 stage 4 (Porewan stadial), and the Last Glacial Maximum of marine &180 stage 2 (Ohakean), the north-eastern ring plain aggraded rapidly with deposition of thick continuous diamicton sequences. The other recognised cool period in the southern North Island, the stadial of late &180 stage 3 (Ratan}, did not appear to induce major aggradation on the north-eastern ring plain. During periods of mild climate within the Last Glacial, deposition on the north-eastern ring plain was dominated by fa ll accession of either tephra, or material reworked from other parts of the ring plain by aeolian processes.

7 .3.1 Introduction

Ruapehu volcano is an active andesitic composite volcano located at the southern end of the Taupe Volcanic Zone (TVZ) in the central North Island of New Zealand. Previous studies of the geological history of this volcano have mainly been confined to the present volcanic cone (e.g. Gregg, 1960; Hackett and Houghton, 1989), or to the younger (<22.5 ka) stratigraphy preserved on the ring plain, where a more complete record of events at the volcano are recorded, both spatially and temporally (e.g. Pal mer, 1991; Donoghue et al., 1995). This paper presents a summary of the stratigraphy preserved in the north-eastern sector of the Ruapehu ring plain and describes events occurring on the volcano and this sector of the ring plain for the period 22.5 ka to c. 80 ka.

7.3.2 Geologic setting

The Tongariro Volcanic Centre (TgVC) at the southern terminus of TVZ is composed of five composite andesitic volcanoes, as well as a number of outlying smaller vents. Ruapehu volcano is the largest in the TgVC and at 2797 m is the highest peak in the North Island. Tongariro volcano is also a large massif which lies to the north of Ruapehu. Both volcanoes have been historically active. The current Ruapehu massif comprises a volume of 110 km3 (Hackett and Houghton, 1989) with a surrounding volcaniclastic ring plain of similar volume. Ruapehu is directly underlain by Tertiary marine sediments with Mesozoic metamorphosed greywacke (schist) occurring at greater depths. The Ruapehu ring plain deposits extend down drainage channels into Tertiary hill country to the south and west, interfinger with the Tongariro ring plain to the north, and are confined in the east by the Kaimanawa Mountains.

155 7 .3.3 Site of study

The Upper Waikato Stream drains the north-eastern flanks of Ruapehu volcano and is immediately north of the watershed boundary where the present drainage from the eastern flanks of Ruapehu divides between the Whangaehu River to the south and the Tongariro River system to the north (Fig. 7.3). Between State Highway 1 and the Tongariro River, Upper Waikato Stream has cut deeply through the ring plain deposits of Ruapehu volcano, exposing in near continuous sections a record of deposition spanning an estimated 80 000 years.

7.3.4 Stratigraphy

The stratigraphy of the deposits in the age range of 22.5 ka to the present has been described by Topping (1973) and Donoghue (1995). Beyond 22.5 ka there is very little published information on the Ruapehu and Tongariro ring plains. Topping (1973) describes these older deposits as the Rangipo lahars on the Tongariro ring plain, and the Waimarino lahars on the Ruapehu ring plain. This study focuses on the deposits older than the regional marker horizon of the Kawakawa Tephra, dated at 22 590 230 years ± BP (Wilson et al., 1988). The composite stratigraphic sequence preserved in Upper Waikato Stream cuttings beneath the Kawakawa Tephra is shown in Fig. 7.4.

7.3.4A Age Time control is provided by the presence of five distal rhyolitic tephra layers interbedded with sediments in the depositional sequence. These are identified here as the Taupo sourced Kawakawa and Okaia Tephras, and the Okataina Volcano-sourced Omataroa Tephra, Hauparu Tephra, and Rotoehu Ash (Howorth, 1975; Froggatt and Lowe, 1990). The ages of these units are given in Table 7.2, and their stratigraphic position in Fig. 7.4.

Table 7.2 Rhyolitic tephra identified within the northeastern ring plain sequence, Ruapehu. Tephra name Source * Age (B.P.) Reference for age

Kawakawa Tephra TVC 22 590 230t Wilson et al. (1988) ± Okaia Tephra TVC ea. 23 000* Froggatt and Lowe (1990) Omataroa Tephra ovc 28 220 630t ± Hauparu Tephra ovc 35 870 1270t ± Rotoehu Ash ovc 64 000 4000§ Wilson et al. (1992) ± • TVC - Taupo Volcanic Centre; OVC = Okataina Volcanic Centre. Denotes 14C ages on old half life basis. t * Estimated stratigraphic age. s Whole rock K-Ar age.

The Kawakawa Tephra occurs in this area as a 0.2 to 0.5 m deposit. Abundant accretionary lapilli are contained within the fall layers of the deposit, characteristic of this

156 tephra (Self and Healy, 1987). The older four tephras do not occur as visible layers in the sequence. They were identified as thin zones of rhyolitic glass enrichment within fine grained andesitic ash. The identification of these microscopic tephra layers was elucidated using a combination of major element glass chemistry, ferromagnesian mineralogy, and stratigraphy. The major element chemistry was determined for polished individual grains on a JEOL-733 electron microprobe following the procedures and analytical conditions of Froggatt and Gosson (1982), and Froggatt (1983). A 20 f.lm beam diameter was used when possible, otherwise a 10 f.lm beam was used with a beam current of 8 nA at an accelerating voltage of 15 kV. Between 10 and 20 individual glass shards were analysed for each tephra layer. The glass chemistry was compared with published glass analyses carried out under the same analytical conditions (Froggatt, 1983; Froggatt and Rogers, 1990; Pillans and Wright, 1992; Stokes et al. , 1992; and Pillans et al. , 1993). Similarity coefficients, and coefficients of variation (Borchardt et al. , 1971) were calculated to compare the unknown samples with large volume tephra units in the general time frame. The glass chemistry provided good matches with stratigraphically consistent correlations. The presence of cummingtonite within the layer identified as Rotoehu Ash gave further evidence of its identification based on glass chemistry (Froggatt and Lowe, 1990). The time planes provided by the rhyolitic tephra horizons enable a reconstruction of the events building up the north-eastern sector of Ruapehu ring plain. Near the base of the sequence a lignite deposit is exposed. Given the time control provided by the rhyolitic tephras above, this lignite may be equivalent in age to one of two other lignite deposits found adjacent to this area. McGione and Topping (1983) describe two lignite deposits each representing warm periods during marine 8180 stage 5a and 5c. Further palynology work will confirm whether the lignite deposit can be correlated to one of these warm periods at approximately 80 ka and 1 OOka respectively (See Section 7.3.9).

7.3.48 Sequence The stratigraphic column is divided into three parts (Fig. 7.4), each representing periods of different types of deposition on the north-eastern sector of the ring plain. The oldest part of the sequence (column C in Fig. 7.4) contains a cemented diamicton unit at the base. This unit forms the valley floor and its base is not seen. Overlying this basal diamicton is a deposit of dark brown and black, fibrous, firm, and centimetre to decimetre scale laminar-bedded lignite. Interbedded within this lignite, as well as immediately above and below, are hornblende-rich andesitic, lapilli and ash layers. These tephra layers vary in thickness from centimetres to tens of centimetres and are not exposed in many other localities; hence their distribution is poorly known. Interbedded within the lignite, as well as above it, are centimetre-scale finely-bedded fluvial sands. Above the lignite deposit the remainder of column C is composed of stacked grey metre-thick diamictons as well as bedded sands, silts, and gravels.

157 175° 45'

Mt To ngariro .A Mt Ng�)ho� 1500 m

39° 10 I

• = Section locations

0 5km - State Highways Figure 7.3 Location of To ngariro Volcanic Centre including Mts. To ngariro, Ngauruhoe and Ruapehu, and the streams dissecting the eastern Ruapehu and To ngariro ring plains. Upper Waikato Stream section locations marked by dots.

158 A B C (m) (m) (m) 0 Section Key Okaia Te phra

Kawakawa Te phra Omataroa Te phra / - 1 D grey andesitic tuft - A �{Jl 1 � 2 D 10�...... d 3 0- -- ��i:�� B I �.,P:. � ). ... I 4 0

01 CD ...... - c 15 ... . 5 0 ····· ···I I 30

Roto Ash ,____ . 7 - Lignite Ce1ment ed � Sandy matrix debris diamicton flow deposits Hyperconcentrated N,....,.,.,"< ,...,..;,···�"""·�{l streamflow deposits 20-n � Silty matrix debris '· ;>,.t' · ·r....1 flow deposits 3 Lignite Fine andesitic ash Andesitic lapilli layers i Strongest paleosol development �'""� ·'""'> :...,.-:"'""'·:-:'""-"1 Bedded sands, silts and gravels

Figure 7.4 Composite stratigraphic columns of the northeastern Ruapehu ring plain sequence below the regional marker horizon of the Kawakawa Te phra. These deposits are unconsolidated and there is no evidence for significant pause between the emplacement of each unit. The mechanisms of emplacement range from stream flow through hyperconcentrated flow and debris flow. There are a few andesitic lapilli layers interbedded near the top of this part of the sequence. Above the stacked diamictons in column C the nature of the deposits, and hence the style of deposition changes markedly. Column B shows that deposits on this sector of the ring plain now consist dominantly of andesitic pumice lapilli and fine ash. The occasional diamicton unit still occurs but these appear to be separated by significant thicknesses of lapilli and fine ash. The cemented diamicton in the central part of column B contains up to 50% by volume pumice clasts and a pumice lapilli layer occurs between two of the flow units of this deposit. At the base of column B the sequence consists of many andesitic pumice lapilli layers stacked on top of one another with very little fine ash between each layer. Further up the column there are greater thicknesses of fine ash material between each pumice lapilli unit. The ash is brown to yellowish brown in colour, loam to clay loam in texture, very firm in consistence, and has a coarse blocky soil structure when dry. Root rhizomorphs are present throughout the ash and these occur in the greatest concentration where the ash is darker in colour and has more evidence of soil structure and soil development (indicated on column B as the zones of strong weathering). The grey andesitic tuff unit (Fig. 7.4) is firmly cemented, shower-bedded and contains accretionary lapilli. This unit is a locally important marker horizon. Column A shows the upper portion of the sequence, the nature of deposition inferred to have again changed. At this time deposition on this part of the ring plain consisted of stacked diamictons of debris, hyperconcentrated, and stream flow origin. There appears to have been little or no time between depositional events. Thick (0.5 m) andesitic lapilli units are interbedded between and within the diamictons and fluvial deposits (column A, Fig. 7.4). Above the Kawakawa Tephra, fluvial aggradation deposits of the Hinuera Formation are described by Topping and Kohn (1973) in the Ruapehu and Tongariro area.

7.3.5 Lithology

7 .3.5A Diamictons Volumetrically this sequence is dominated by diamicton deposits. These vary considerably in their nature and inferred style of deposition. At one end of the depositional series are bedded silts, sands, and gravels. These are well sorted within individual layers but are poorly sorted as a whole, with layers of differing grain size bedded on top of one another. These deposits reflect a stream flow origin where turbulent, dilute (Newtonian) flow mechanisms transport the sediment (Smith, 1986; Pierson and Costa, 1987). At the other end of the scale are debris flow deposits which are poorly sorted and display no bedding. Often these deposits contain very large clasts within a fine sand matrix. These deposits have been transported and deposited by a much more concentrated (plastic or non-Newtonian) flow mechanism. Laminar flow

160 occurs under these conditions and sediment is supported and transported by a strong matrix (Smith, 1986; Pierson and Costa, 1987). Between these two extremes of deposit are a wide range of deposits intermediate in character. These deposits show varying amounts of planar bedding and varying degrees of sorting, with generally a finer grain size than debris flows. These units are termed hyperconcentrated flow deposits and are considered to have been deposited via an intermediate concentration (but still plastic and non-Newtonian) flow mechanism. Turbulence and matrix strength both play a role in transporting and depositing sediment in these flows (Smith, 1986; Pierson and Costa, 1987). There exists a wide range of deposits transitional in character between these three types. Table 7.3 outlines the characteristic features of the surface flow sediments within the northeastern ring plain sequence, and relates the observed (purely descriptive) types of deposits with their features and inferred flow mechanisms.

Table 7.3 Surface-flow sediment lithotypes within the northeastern Ruapehu ring plain area, with their characteristic properties and inferred mode of deposition. Sediment type Observedfeatures Inferred flow type(S mith, 1986) Cemented - fi rmly cemented Debris flow and diamictons - sand matrix hyperconcentrated flow - matrix- and clast-supported - pumice lapilli rich (up to 50% by vol.) - maximum clast diameter 1-3 m - composed of several flow units - massive and planar fabric preserved

Sandy matrix - unconsolidated Debris flow, hyperconcentrated diamictons - fine-coarse sand matrix flow, and transitional to stream - matrix- and clast-supported flow - pumice lapilli poor, dominated by lithic clasts - maximum clast diameter 1 m - massive and planar fabric preserved

Silty matrix - unconsolidated Debris flow diamictons -greasy, allophane rich, silt loam matrix - matrix-supported - maximum clast diameter 2 m - can contain hydrothermally altered clasts

Bedded silts and - unconsolidated, dominantly clast-supported Stream flow and transitional to sands, and - planar laminated and low angle cross-bedded hyperconcentrated flow gravels - pumice and lithic rich zones - highly variable grainsize, silt, sand, and pebbles in well sorted layersand lenses - maximum clast diameter 0.2m

7.3.58 Andesitic Tephra Andesitic tephra deposits are common throughout the entire sequence. More than 60 individual lapilli-grade units are described in the sequence, and thick deposits of fine ash occur in parts of it. The andesitic lapilli units are dominantly pumice that is highly vesicular with very fine vesicles, intermixed with <1 0% wall-rock lithic lapilli. The lapilli layers are generally thickest immediately to the east of Ruapehu volcano, which is consistent with the prevailing westerly wind direction. Close to source there is no

161 evidence of these tephra layers and their preservation on the ring plain is patchy, thus the best way to describe their original distribution and volume is by comparison to better studied younger tephra layers of the Bullot Formation (Donoghue et a/., 1995). Several analogous layers can be found which give an idea of the approximate dimensions of these eruptives. The lapilli layers would appear to be mostly of a sub-plinian origin, although plinian eruptives cannot be ruled out as the dispersion of the deposits is not well known. The tephra contain phenocrysts of plagioclase + orthopyroxene + clinopyroxene > titanomagnetite ± olivine ± hornblende. This assemblage accounts for 90% of the andesitic units in the sequence. Hornblende and olivine occur only in small quantities (<2% by volume) in most of the units. Two other ferromagnesian mineral assemblages are observed in the andesitic tephra which can be characterised as, hornblende-dominant or olivine-dominant assemblages: Olivine > Orthopyroxene Clinopyroxene > Titanomagnetite + Hornblende > Orthopyroxene + Clinopyroxene > Titanomagnetite These assemblages are restricted in their occurrence within the sequence. Two tephra layers at 12 m depth in the sequence (Column A, Fig. 7.4) have an olivine-dominated ferromagnesian assemblage, while two tephra layers at the very base of the sequence within lignite (Column C, Fig 7.4) contain a hornblende dominated assemblage. Andesitic lapilli falls occur throughout the sequence, and the rate of eruption is > one sub-plinian lapilli eruption per 1000 years for a period of approximately 60 000 years. Thick deposits of fine andesitic ash also occur in parts of the sequence and these are thought to represent the slow accumulation of fine ash directly from many eruptions as well as wind-blown fine ash reworked from other parts of the ring plain.

7.3.6 Interpretation of geological history

The relationship between the interpreted geological history of the north-eastern ring plain, the 8180 record of deep sea cores, and the record of loess deposition and paleosol development of the southern North Island is summarised in Fig. 7.5. The lignite in the lower part of the sequence is thought to represent deposition during marine 8180 stage 5a, the Otamangakau interstadial period of McGione and Topping (1983) around 80 ka (See Section 7.3.9). The organic materials accumulated slowly in a bog with occasional deposition of fluvial fine sands, and hornblende-rich tephra falls. The source of these hornblende-rich tephras could be either Tongariro or Ruapehu volcano, both of which were potentially active at the time. No hornblende-rich tephras have yet been described from Ruapehu whereas younger Tongariro-sourced tephras occasionally contain appreciable hornblende (Lowe, 1988; Donoghue et al., 1991; Donoghue et a/., 1995). This suggests that Tongariro Volcano is the most likely source for these tephras. The stacked diamictons and bedded silts, sands and gravels above the lignite represent a period of rapid aggradation on the north-eastern sector of the ring plain.

162 Aggradation without significant pause would imply that the slopes of Ruapehu were unstable, with much loose rock material available for transport and deposition to the north-east by lahars and stream flow. There are few interbedded lapilli units within these sediments and also little primary pumice within the sediments themselves. This indicates that these aggradation deposits are not likely to be the result of large scale eruptive events on Ruapehu volcano. This does not rule out the role of small scale eruptive events which are not recorded in the stratigraphic record as tephra deposits, but which have the potential to produce lahar and flood depositional events. A good example of this were the lahars produced during the eruptions of Ruapehu in 1975 (Nairn et al., 1979), an eruption which was not accompanied by a tephra unit preserved on the ring plain. Small scale eruptives could potentially have produced the stack of diamictons and bedded deposits described here, but it is here considered that other environmental factors also played a major part. Aggradation was taking place prior to the eruption of the Rotoehu Ash at ca.64 ka (Wilson et al., 1992) and occurred during marine 8180 stage 4 (Porewan stadial). The cooler climate of this period possibly accelerated physical weathering processes on Ruapehu, thus providing abundant material for redeposition on the north-eastern ring plain. The mechanism of this redeposition may have been small scale eruptives or simply slope failures and floods during storm events. The presence of a greater area of snow and ice on the higher slopes of the volcano in the cooler climate of 818 isotope stage 4 may have also provided a ready water supply for producing lahars and floods in small scale eruptions, such as was observed in the 1985 eruption of Nevado del Ruiz (Pierson et al., 1990). Above the stacked diamictons, the deposits on this sector of the ring plain consist dominantly of andesitic pumice lapilli and fine ash beds. Occasional diamicton units occur but appear to be separated by significant periods of time and hence do not suggest major instability on the volcano. Deposition of the pumice-rich cemented diamicton unit (in the central part of column 8, Fig. 7.4), which contains an andesitic pumice lapilli layer between flow units, probably was initiated by a large tephra-producing eruption on Ruapehu volcano. At the base of the sequence represented by column B, the eruption rate of pumice lapilli falls on this sector of the ring plain appear to have been very high because there is little, or no fine ash between each lapilli layer and the lapilli units are often very thick and stacked on top of one another. Such a high eruption rate occurred in late marine 8180 stage 4, corresponding to the period encompassed by the previously-named middle Tongariro Subgroup tephras (Leamy et al. , 1973; Milne and Smalley, 1979) which are mapped as occurring on top of the Porewa loess in the Rangitikei region to the south of Mt Ruapehu. Distal andesitic ash is also mapped on top of a correlative of the Porewa loess in the Hawkes Bay region to the east of Mt. Ruapehu (A.P. Hammond, pers. comm., 1994 ). Thus the lapilli layers found on the north-eastern ring plain are probably proximal correlatives of the middle Tongariro Subgroup tephras.

163 Age 0 Isotope Rhyolitic NE Ruapehu Rangitikei Wanganui Taranaki (ka) stages tephras ring plain

20 2 Oh L1 Sy1

------1 1 Soil development I Whanahuia and tephra accession Paleosol 30 1111111111 lll�i;ll ll11111's'r2 11

Hauparu ------+--� Rt 40 Sy2 Te phra accession L2 ...... 3 0> � 50 Kimbolton Soil development Paleosol and tephra accession 60 Rotoe ------1 1111r1 11111 1111fr�� 11 hu ''' '''' '� ______Middle To ngariro ______I'''''''''I ------Wttttm Subgroup 70 4 Fluvial and lahar Po L3 Sy3 aggradation 11111 Tapuae 11111 5a Peat accumulation Sr4 80 Paleosol lillt lLwl Figure 7.5 Comparison of the northeastern Ruapehu ring-plain landscape events with deep sea 0 isotope record (Shakleton et al., 1990), rhyolitic tephra marker beds (Froggatt and Lowe, 1990), and southern North Island loess stratigraphy from the Rangitikei valley (Leamy et al., 1973; Milne and Smalley, 1979), Wanganui (Pillans, 1988; 1994), and Taranaki (AIIoway et al., 1988). Oh = Ohakea loess, Rt = Rata loess and Po = Porewa loess, L 1 = loess 1, S2 = buried soil 2 etc. Sy1 = loess 1 correlative, Sr2 = buried soil 2 correaltive etc. Above these stacked pumice lapilli beds, the eruption rate of lapilli grade units appears to have slowed, indicated by greater thicknesses of fine ash between each lapilli unit. This may indicate that the dispersal axes of eruptives changed at this time or there was a reduction in magnitude or frequency of eruptive events. Given that the sites studied are directly down wind of the volcano under the prevailing westerly winds, it is unlikely that eruptives were deposited in different directions for the extended periods of time represented by this sequence. Therefore it is more likely that there was a reduction in the magnitude and/or the · frequency of eruptive events at this time. Coupled with this apparent slowing of eruption rate or reduction in magnitude of the eruptive events, the degree of weathering and paleosol development within the fine ash increases, as the soil surface was accreting more slowly. The strongly weathered zone extends to just above the level of the Rotoehu Ash. This paleosol appears to have formed at the same time as the paleosol developed in the Porewa loess throughout the southern North Island (Leamy et al., 1973; Milne and Smalley, 1979; Palmer, 1985; Pillans, 1988) during early marine 8160 stage 3. During the subsequent stadial period in mid-marine 8160 stage 3 (Ratan) there is no expression of major aggradation on the north-eastern ring plain. The diamicton bracketed by the Hauparu and Omataroa tephras contains andesitic lapilli and appears to have been generated by eruptive rather than climatic processes, in contrast to the earlier diamictons. Deposition continued with accretion of fine ash and pumice lapilli beds. The weathering and soil development within fine ash deposited in this period is, however, very weak. This indicates that, either the eruption or accumulation rate was too high to allow soil development, or soil development was affected by cooler climate conditions during this period. The fine ash nature of the deposits and the thickness of ash between dated rhyolitic tephra layers would suggest a relatively slow deposition rate, indicating soil development was probably slowed by cool climate conditions. The absence of appreciable aggradation deposits on the north-eastern sector of the ring plain at this time may indicate that the climate during the mid-marine 8160 stage 3 stadial was not as severe as during the marine 8160 stage 4 stadial. This is supported by pollen evidence from the area. McGione and Topping (1983) and McGione (1985) proposed a "cool interstadial" climate in the Ruapehu area during the time of the deposition of the Rata loess rather than a full stadia! period. In Taranaki, andic deposits record peaks of aerosolic quartz accumulation indicative of cool climate in marine 8160 stages 2 and 4, but negligible amounts in stage 3 {AIIoway et al., 1992). This also indicates the climate was not as severe in marine 8160 stage 3 in comparison with that of stages 2 and 4. Above the grey tuff unit in column B, the fine ash exhibits much stronger weathering. Root rhizomorphs become abundant and finely-disseminated charcoal layers are present. Near the top of this strongly weathered zone is the Omataroa Tephra, dated at 28 220 630 BP (Froggatt and Lowe, 1990). This paleosol development correlates ± with the paleosol in the Rata loess in the southern North Island (Leamy et al., 1973; Milne and Smalley, 1979; Palmer, 1985; Pillans, 1988), developed during late marine 8160 stage

165 3. The fine ash in this part of the sequence is not interbedded with any lapilli layers and appears to have accumulated slowly, thus enhancing the paleosol and weathering features. Weathering features become less pronounced above the Omataroa Tephra, potentially indicating deteriorating climate with the onset of 8180 stage 2 (Last Glacial Maximum; Pillans et al., 1993). Above the Okaia Tephra, rapid aggradation of diamictons began on the north eastern ring plain, again indicating instability on the slopes of Ruapehu volcano. These deposits accumulated during marine 8180 stage 2. Throughout the aggradation sequence, represented by column A, andesitic lapilli continued to be erupted, indicating Ruapehu was active throughout the early part of the Last Glacial Maximum.

7.3.7 Discussion and Conclusions

The north-eastern Ruapehu ring plain sequence not only reflects the volcanic history of Ruapehu volcano but also indicates the response of the volcano-ring plain system to global climate oscillations. Periods of severe-cold climate during the Last Glacial have caused large scale aggradation on the ring plain, with more frequent occurrences of lahars and deposition dominated by surface flow mechanisms. These processes appear to attain their maximum volume and frequency during marine 8180 stages 2 and 4 (Fig. 7.5). The climate during mid-marine 8180 stage 3 may not have been so severe, because no widespread aggradation is noted in the deposits of this age on the north-eastern ring plain. The climate-related aggradation may be driven by increased physical weathering on the higher slopes of Ruapehu volcano. The aggradation deposits to some degree mask the purely eruptive or volcanic triggered events recorded on the ring plain. Only deposits from the very large tephra eruptions are preserved in this environment; the smaller tephra units and fine ash deposits are mostly reworked by the wind or entrained in successive lahars and stream flow deposits. In the milder periods on the north-eastern ring plain, deposition reflects the eruptive and volcanic-triggered events occurring on Ruapehu volcano. Deposition is dominated by fall processes, primary fall ash and lapilli deposition and aeolian redistribution of fine ash on the ring plain. Diamicton deposits are still present in this sector but the time between each lahar event is much greater than during 8180 stages 2 and 4.

7 .3.8 Acknowledgements

This work forms part of the Ph.D. research of one of the authors (SJC). The authors gratefully acknowledge funding from the New Zealand Vice-Chancellors' Committee, Massey University Graduate Research Fund, the Helen E. Akers Scholarship Fund, and

166 the Department of Soil Science of Massey University. Thanks are also due to B.V. Alloway, W.R. Hackett, and B.F. Houghton for their thoughtful and insightful reviews, which greatly improved the manuscript.

7.3.9 Subsequent comments on the paper

A recent analysis of the palynoflora within the lignite deposit at the base of the Upper Waikato Stream sequence has led to a slight modification of its estimated age. The lignite contains a palynoflora composed of shrub and grassland species, indicative of 1 a cool and stormy climate (McGione, pers. comm., 1995). Thus its correlation to & 80 stage 5a in the paper (a relatively warm period) is not supported. Instead it is now considered (see Chapter 5) that the lignite was deposited immediately after &180 stage 5a, within &180 stage 4 (or the Porewan stadial of the Last Glacial). Consequently the record preserved on the northeastern ring plain extends back only to the Porewan stadial (represented by the Porewan loess, Loess 3 and Sr3 on Fig. 7.5). In addition, the identification methods of the interbedded rhyolitic tephras have been improved over those described in the paper (see Chapter 2).

7.4 SUMMARY GEOLOGICAL SYNTHESIS OF THE NORTHEASTERN TONGARIRO VOLCANIC CENTRE

This synthesis concentrates on the events which occurred on the ring plains rather than on the Tongariro and Ruapehu volcano slopes.

7.4.1 c. 75 000 -64 000 years ago

The record from this period is preserved only in the Upper Waikato Stream sequences described in Section 7.3 and 5.13 (Chapter 5). This interval encompassed the antepenultimate stadial of the Last Glacial (the Porewan stadial), thus the periglacial climate of this time had a large influence on ring plain sedimentation. Palynoflora within a lignite deposit from this sequence indicates that the climate was cool and stormy during this time (Section 5.13, and 7.3.9). The northeastern Ruapehu ring plain during this interval was aggrading rapidly with deposits from many lahars across it. These lahars were mostly non-cohesive, sandy matrix flows exhibiting both hyperconcentrated streamflow and debris flow sedimentation styles. Reworking of these lahar deposits probably occurred soon after their emplacement and large thicknesses of alluvial gravelly sands are associated with many of the lahar deposits. The lahars during this time were likely to be in response to greater physical weathering on the volcano slopes and consequently larger loads of sediment within the upper stream catchments. Glaciers were probably present on the cone and supplied sediment (from till and outwash) and water (from ice melt and kettle lakes) to lahars. Lahar-triggering mechanisms probably included; eruptions from either Ruapehu

167 or Tongariro, storms, glacial lake breakouts, avalanches and possibly small flank collapses. During the main period of ring-plain aggradation by lahars, the frequency of large tephra eruptions from either Ruapehu or Tongariro appears to have been low. However, hornblende-rich tephras below the lahar deposit sequence may indicate that in the early part of this interval Tongariro volcano was actively erupting (Section 3.4.7 and 7.3.6). In addition, following the main period of lahar aggradation, but prior to 64 ka, it appears that there was a major period of eruptive activity from Ruapehu volcano (Fig. 7.6). Large numbers of thick lapilli layers interbedded within fine andesitic ash, indicate that large and frequent eruptions were occurring at Ruapehu {Table 5.3, Sections 4.10 and 7.3.6). Fewer lahar deposits preserved in this part of the sequence imply that the climate had begun to ameliorate and the volcano slopes were more stable, despite these frequent and large tephra eruptions.

al 5 ro E :;:::. 1/) UJ 0 80+---��----��----�-L� 60 40 20 0 Time range (ka)

Figure 7.6. Summary diagram of the estimated rate (per ka) of lapilli-producing sub plinian eruptions of Ruapehu and Tongariro volcanoes for the last 75 ka. Based on the ring plain record of Topping (1973), Donoghue et al., (1995) and Cronin et al., (1996).

7 .4.2 64 000 - 36 000 years ago

The most complete record of this interval is also preserved in the Upper Waikato Stream sequences. During this period, very little lahar deposition occurred on the northeastern Ruapehu ring plain. The deposits of only a single lahar are preserved at only a few localities (R13, Section 5.9). This was probably related to stabilised landscapes in the

168 mild and settled climatic conditions prevailing during this interval in the area (McGione and Topping, 1983). A very low frequency (1 per 7 000 years) of large tephra eruptions from Ruapehu was recorded in this period (Fig. 7.6, Table 5.3, Sections 4.10 and 7.3.6). However, accretion of fine ash appears to have occurred throughout the interval, indicating frequent low-magnitude eruptions, from either Ruapehu or Tongariro. Soil development indicates that during this interval, the northeastern Ruapehu ring plain was a stable landscape surface, with low rates of accretion (Section 4.1 0). Stronger soil development in the earlier part of this interval corresponded in age to soil development in loess sequences throughout the lower North Island (Sections 4.10 and 7.3.6), during a period of warmer and more settled climate.

7 .4.3 36 000 - 23 000 years ago

The full record of this period is also preserved only in the Upper Waikato Stream sequences. As in the preceding interval, deposition on the northeastern Ruapehu ring plain was dominated by accession of tephras. However, two lahars, closely spaced in time, inundated the area. These lahars were contemporaneous with and probably related to large scale tephra eruptions of Ruapehu (R12, Sections 5.9 and 5.13). Large-scale tephra eruptions of Ruapehu were much more frequent in this period compared to the last, with an average of 1 per 800 years (Fig. 7.6, Table 5.3, Sections 4.10 and 7 .3.6). However, the overall frequency of large scale eruptions was not uniform throughout the entire interval. The rate was high early in the period, but very low in the latter stages. Smaller scale eruptions (probably from both Ruapehu and Tongariro) occurred throughout the entire interval and resulted in continuous and slow accretion of the ring-plain surface. Soil development was strongest in the latter stages of the 36 - 23 ka interval due to the rate of large scale tephra eruptions being at their lowest, thus the soil surface was accreting slowly. This period of strong soil development corresponds to a period of widespread soil development in loess sequences throughout the lower North Island (Section 4.1 0).

7 .4.4 23 000 - 15 000 years ago

This was a period of major construction of both the Tongariro and Ruapehu ring plains due to both volcanic activity and climatic instability. On both ring plains deposits of this interval probably comprise a larger volume than those of any other interval described. This interval corresponds with the period of greatest climatic and landscape instability in the North Island over the last 120 ka (Pillans, et al. 1993). Response to this cooler and stormier climate on the ring plains was deposition of large amounts of sediment by lahars and streams (Section 5.13). Thick deposits of many lahars were deposited over the entire ring plain study area (R1 1, R1 0 and R09 on the Ruapehu ring plain, T3 and T4 on

169 the Tongariro ring plain; Section 5.12, Fig 5.7). In addition to lahar sedimentation, all of the streams in the area were transporting and depositing large quantities of sediment as well as reworking lahar deposits. On the Tongariro ring plain, particularly in its northern portion, emplacement of the Oruanui lgnimbrite contributed large volumes of material which was reworked and preserved extensively as Hinuera Formation alluvium (Sections 5.11 and 5.13) and remobilised by wind as Mokai Sand. Eruptive activity of Ruapehu volcano in this interval was also very high, contributing to construction of the ring plains by large additions of tephra as well as generation of lahars. The voluminous middle and lower Bullot Formation tephras from Ruapehu volcano were erupted between 22.6 and 14.7 ka B.P. (Donoghue et al., 1995). In addition, between the eruptions of Okaia and Kawakawa Tephras (22.6 ka B.P. to ea. 23 ka) several more thick pumice lapilli layers were erupted and preserved between lahar and fluvial deposits (Fig. 7.6, Section 3.4, Fig. 3.6). Soil development in this interval was minimal due to climatic instability (i.e. scant vegetation cover) and rapid aggradation of ring plain surfaces with lahar, fluvial and pumice tephra deposits. A large lava flow from Tongariro volcano occurred in the early part of this interval. This flow was emplaced between the ring plains of Ruapehu and Tongariro creating a boundary between them (Section 5.12). Other lava flows on the northeastern Ruapehu ring plain may have also been emplaced during this interval (Section 7.2.3A).

7 .4.5 15 000 - 9 700 years ago

In this interval the high rate of large tephra eruptions from Ruapehu volcano continued with large thicknesses of upper Bullot Formation tephras deposited on the northeastern Ruapehu and southeastern Tongariro ring plains (Donoghue et al. , 1995). Tongariro volcano was also extremely active in this period, with the eruptions of the Rotoaira, Pahoka and Mangamate Tephras (Topping, 1973). The maximum rate of large lapilli producing eruptions was between 10 and 9.7 ka B.P. (Fig. 7.6), when at least six sub plinian tephras were erupted from Tongariro (Mangamate Tephra). Many lahars continued to occur throughout this interval, although they were confined to narrow sectors of the ring plains. On the Ruapehu ring plain, lahar deposition was centred around the Mangatoetoenui - Te Piripiri Streams area and on the Tongariro ring plain, around the Oturere - Makahikatoa Streams (ROB - R02 and T1 , Section 5.12). These lahars were related to the frequent tephra eruptions of both volcanoes in this interval (Section 5.13). The major portion of the ring plains in between were unaffected by lahar and fluvial deposition and had relatively stable surfaces showing periods of soil development punctuated by deposition of thick tephra layers. The absence of widespread fluvial and lahar aggradation of the ring plains in this interval compared to the last, corresponds with post Otiran climatic amelioration (McGione and Topping, 1977). During this interval, lahars caused major aggradation in the lower Tongariro River, resulting in extensive laharic surfaces. These deposits ceased accumulating at around 9 700 years ago (Section 6.9.2A).

170 Another event that occurred during this interval was between 14.7 and 11.9 ka B.P., when a lava flow was emplaced on the northern flanks of Tongariro, reaching the ring plain (Section 7.2.38).

7 .4 . 6 9 700 - 2 500 years ago

During this interval most of the surfaces of both ring plains were stable with slow accretion of tephra from small eruptions of Ruapehu and Tongariro forming the Papakai Formation (Topping, 1973; Donoghue et al., 1995). Soil development was strong throughout this interval with the combination of a mild climate (McGione and Topping, 1977) and slow soil surface accretion. A single lahar occurred during this period but deposition was confined close to the Mangatoetoenui Stream (R01, Section 5.12). This lahar was probably related to a small eruption of Ruapehu (Section 5.13). In the Tongariro River the extensive laharic surface constructed in the preceding interval was incised and in places a small laharic surface was constructed at ea. 5.2 ka (Section 6.9.28 and 6.1 0.4 ).

7 . 4 . 7 2 500 years ago - present

Tephra accretion and soil development continued throughout this interval with deposition of the Mangatawai Tephra, Tufa Trig Formation and Ngauruhoe Formation (Topping, 1973; Donoghue et al., 1995). However, this was interrupted by the emplacement of large volumes of Taupo lgnimbrite. The lgnimbrite was up to 30 m thick in the Tongariro River valley and in many places on the ring plain, levelling surfaces of formerly differing ages (Section 6.9.2A). In the Tongariro River incision continued, particularly following infilling of the river valley with Taupo lgnimbrite. At least seven small lahars occurred in this interval, some producing depositional surfaces in the lower Tongariro River, others documented only in historic records and two in 1995 which left deposits along the Mangatoetoenui Stream (6.9.28). All of the pre-1995 lahars in this interval appear to have flowed down the Mangatoetoenui Stream. These lahars were all probably related to eruptions of Ruapehu.

7.5 Combined References

Alloway, 8.V., Stewart, R.8., Neall, V.E. and Vucetich, C.G., 1992. Climate of the Last Glaciation in New Zealand, based on aerosolic quartz influx in an andesitic terrain. Quat. Res., 38: 170-179. 8orchardt, G.A., Harward, M.E. and Schmitt, R.A., 1971. Correlation of ash deposits by activation analysis of glass separates. Quat. Res., 1: 247-260. Cronin, S.J., Neall, V.E., Stewart, R.8., Palmer, A.S., 1996. A multiple-parameter approach to andesitic tephra correlation, Ruapehu volcano, New Zealand. J. Volcano!. and Geotherm. Res., 72: 199-215.

171 Donoghue, S.L., 1991 . Late Quaternary volcanic stratigraphy of the south-eastern sector of Mount Ruapehu ring plain, New Zealand. Unpub. PhD. thesis, Massey University, Palmerston North, N. Z. Donoghue, S.L. , Neall, V.E., Palmer, A.S., 1995. Stratigraphy and chronology of andesitic tephra deposits, Tongariro Volcanic Centre, New Zealand. J. Roy. Soc. N. Z., 25: 115-206. Donoghue, S.L., Stewart, R.B. and Palmer, A.S., 1991. Morphology and chemistry of olivine phenocrysts of Mangamate Tephra, Tongariro Volcanic Centre, New Zealand. J. Roy. Soc. N. Z., 21: 225-236. DOSLI (Department of Survey and Land Information), 1982. NZMS 260, T20 Ruapehu, Edition 1. Wellington N.Z. DOSLI (Department of Survey and Land Information), 1994. Topomap Tongariro. lnfomap 260, T19, Edition 2. Wellington N.Z. Froggatt, P.C. and Gosson, G.J., 1982. Techniques for the preparation of tephra samples for mineral or chemical analysis and radiometric dating. Geology Dept., Victoria University of Wellington Pub I., 23: 12 pp. Froggatt, P.C., 1983. Towards a comprehensive Upper Quaternary tephra and ignimbrite stratigraphy in New Zealand using electron microprobe analysis of glass shards. Quat. Res., 19: 188-200. Froggatt, P.C. and Lowe, D.J., 1990. A review of late Quaternary silicic and some other tephra formations from New Zealand: their stratigraphy, nomenclature, distribution, volume, and age. N. Z. J. Geol and Geophys., 33: 89-109. Froggatt, P.C. and Rogers, G.M., 1990. Tephrostratigraphy of high altitude peat bogs along the axial ranges, North Island, New Zealand. N. Z. J. Geol. and Geophys., 33: 111-124. Graham, I.J., Hackett, W.R., 1987. Petrology of calc-alkaline lavas from Ruapehu volcano and related vents, Taupo Volcanic Zone, New Zealand. J. of Petrology, 28: 531-567. Gregg, D.R., 1960. The geology of Tongariro Subdivision. N. Z. Geological Survey Bull., 40: 152 pp. Grindley, G.W., 1960. Geological map of New Zealand, 1st ed., 1: 250 000, Sheet 8, Taupo. D.S.I.R., Wellington. Hackett, W.R., 1985. Geology and petrology of Ruapehu volcano and related vents. Unpub. thesis Victoria University of Wellington: 312 pp. Hackett, W.R. and Houghton, B.F., 1989. A facies model for a Quaternary andesitic composite volcano: Ruapehu, New Zealand. Bull. Volcano!., 51 : 51-68. Hobden, B. J., Houghton, B.F., Lanphere, M.A. and Nairn, I.A., 1996. Growth of the Tongariro volcanic complex: new evidence from K-Ar age determinations. N.Z. J. Geol. and Geophys., 39: 151-154. Hodgson, K.A., 1993. Late Quaternary lahars from Mount Ruapehu in the Whangaehu River valley. Unpub. PhD. thesis, Massey University, Palmerston North, N.Z.

172 Howorth, R., 1975. New formations of Late Pleistocene tephras from Okataina Volcanic Centre, New Zealand. N. Z. J. Geol. and Geophys., 18: 683-712. Leamy, M.L., Milne, J.D.G., Pullar, W.A. and Bruce, J.G., 1973. Paleopedology and soil stratigraphy in the New Zealand Quaternary succession. N. Z. J. Geol. and Geophys., 16: 723-744. Lowe, D.J., 1988. Stratigraphy, age, composition, and correlation of late Quaternary tephras interbedded with organic sediments in Waikato lakes, North Island, New Zealand., N. Z. J. Geol. and Geophys., 31: 125-165. Mathews, W.H., 1967. A contribution to the geology of the Mount Tongariro massif, North Island, New Zealand. N.Z. J. Geol. and Geophys., 10: 1027-1038. McArthur, J.L. and Shepherd, M.J., 1990. Late Quaternary glaciation of Mt. Ruapehu, North Island, New Zealand. J. Roy. Soc. N. Z., 20: 287-296. McGione, M.S. and Topping, W.W., 1983. Late Quaternary vegetation, Tongariro region, central North Island, New Zealand. N. Z. J. of Botany, 21: 53-76. McGione, M.S., 1985. Biostratigraphy of the Last Interglacial - Glacial cycle, southern North Island, New Zealand. In: Pillans, B.J. (ed.) Proceedings of the second CLIMANZ conference. Geology Dept., Victoria University of Wellington, Publ., 31: 17-31. Milne, J.D.G. and Smalley, I.J., 1979. Loess deposits in the southern part of the North Island of New Zealand: an outline stratigraphy. Acta Geologica Academiae Scientarium Hungaricae, 22: 197-204. Nairn, I.A., Wood, C.P. and Hewson, C.A.Y., 1979. Phreatic eruptions of Ruapehu: April 1975. N. Z. J. Geol. and Geophys., 22: 155-173. Palmer, A.S., 1985. The Last Interglacial record in Wairarapa Valley, New Zealand. In: Pillans B.J. (ed.) Proceedings of the second CLIMANZ conference. Geology Dept., Victoria University of Wellington, Publ., 31 : 32-36. Pal mer, B.A., 1991 . Holocene lahar deposits in the Whakapapa catchment, northwestern ring plain, Ruapehu Volcano (North Island, New Zealand). N. Z. J. Geol. and Geophys., 34: 177-190. Pierson, T. C. and Costa, J. E., 1987. A rheologic classification of subaerial sediment water flows. Geol. Soc. of America, Reviews in Engineering Geology, VII: 1-12. Pierson, T.C., Janda, R.J., Thouret, J-C. and Borrero, C.A., 1990. Perturbation and melting of snow and ice by the 13 November 1985 eruption of Nevado del Ruiz, Colombia, and consequent mobilisation, flow and deposition of lahars. J. Volcano!. and Geotherm. Res., 41: 17-66. Pillans, B.J., 1988. Loess chronology in Wanganui Basin, New Zealand. In: Eden, D.N. and Furkert, R.J. (eds) Loess - Its Distribution Geology and Soils: 175-191. A.A. Balkema, Rotterdam. Pillans, B.J., 1994. Direct marine-terrestrial correlations, Wanganui Basin, New Zealand: The last 1 million years. Quat. Sci. Reviews, 13: 189-200.

173 Pillans, B.J. and Wright, 1., 1992. Late Quaternary tephrostratigraphy from the southern Harve Trough - Bay of Plenty, northeast New Zealand. N. Z. J. Geol. and Geophys., 35: 129-143. Pillans, B., McGione, M., Palmer, A., Mildenhall, D., Alloway, B. and Berger, G., 1993. The Last Glacial Maximum in central and southern North Island: a paleoenvironmental reconstruction using the Kawakawa tephra formation as a chronostratigraphic marker. Palaeo., Palaeo., Palaeo., 101: 283-304. Self, S. and Healy, J., 1987. Wairakei Formation, New Zealand: stratigraphy and correlation. N. Z. J. of Geol. and Geophys., 30: 73-86. Shackleton, N.J., Berger, A. and Peltier, W.R., 1990. An alternative astronomical calibration of the lower Pleistocene timescale based on ODP Site 677. Trans. Roy. Soc. of Edinburgh: Earth Sciences, 81: 251-261. Smith, G.A., 1986. Coarse-grained nonmarine volcaniclastic sediment: Terminology and deposition process. G. S. A. Bull., 97: 1-10. Stokes, S., Lowe, D.J. and Froggatt, P.C., 1992. Discriminant function analysis and correlation of Late Quaternary rhyolitic tephra deposits from Taupo and Okataina volcanoes, New Zealand, using glass shard major element composition. Quat. Inter., 13/14: 103-1 17. Topping, W.W., 1973. Tephrostratigraphy and chronology of late Quaternary eruptives from the Tongariro Volcanic Centre, New Zealand. N. Z. J. Geol. and Geophys., 16: 397-423. Topping, W.W. and Kohn, B.P., 1973. Rhyolitic tephra marker beds in the Tongariro area, North Island, New Zealand. N. Z. J. Geol. and Geophys., 16: 375-395. Wilson, C.J.N., Switzur, R.V. and Ward, A.P., 1988. A new 14C age for the Oruanui (Wairakei) eruption, New Zealand. Geol. Mag., 125: 297-300. Wilson, C.J.N., Houghton, B.F., Lanphere, M.A. and Weaver, S.D. (1992). A new radiometric age estimate for the Rotoehu Ash from Mayor Island volcano, New Zealand. N. Z. J. of Geol. and Geophys., 35: 371 -374.

174 CHAPTER 8: CONCLUSIONS

8.1 Fulfilment of study objectives - Tephrochronology groundwork

Chapter 1 outlined the original objectives of this study as well as identifying areas in which the study could contribute to knowledge gained from past studies in and around the study area. Underpinning all of these objectives was the need to establish a chronostratigraphic framework for the ring-plain sequences. This was eventually provided by a stratigraphy of interbedded andesitic and rhyolitic tephras. For sequences < 22.6 ka B.P. a tephrochronologic framework had been established in past studies (Topping, 1973; Topping and Kohn, 1973; Donoghue et al., 1995). Consequently, in this interval the object of field and laboratory study was to correlate tephras from new sequences to tephras that were already known to occur in the area. For andesitic tephras this was readily achieved using field appearance and stratigraphic criteria. However, rhyolitic tephras were not as easily identified, instead laboratory-based geochemical methods were required. When existing methods were found to be inadequate for precise identification of these rhyolitic tephras, an existing but practically untested statistical method was attempted. This formed the basis of Chapter 2 and resulted in much more precise and quantitative rhyolitic tephra identification. Sequences > 22.6 ka B.P. had not been studied in the area prior to this study and consequently their tephrochronologic framework was unknown. To establish a chronology, rhyolitic tephras were initially sought in the sequences. These were only found after channel sampling and microscopic analysis to identify rhyolitic glass concentration layers within fine-grained andesitic ash sequences. These glass concentrations were correlated with tephras known from the central North Island using the methods described in Chapter 2. Once a rhyolitic tephrochronology was established for this time range in this area it was supplemented by establishment of a new andesitic tephrochronology. Chapter 3 describes the development of the andesitic tephrochronology. The first part of the approach was to test whether the discriminant function analysis (DFA) method used for the rhyolitic tephras in Chapter 2 could be used to discriminate andesitic tephras, and which mineral phases were best for the purpose. This was tested by examining diffe rent methods and geochemistry of different mineral phases to distinguish andesitic tephras from two different sources (Egmont volcano and Ruapehu volcano). Once these methods were proven, they were applied, along with several other parameters, to establish andesitic marker tephras in the > 22.6 ka B.P. sequence on the northeastern Ruapehu ring plain. These marker tephras were then used to correlate the > 22.6 ka B.P. sequence.

175 8.2 Fulfilment of study objectives - Specific objectives

The realisation of the original objectives and planned contributions of this study form Chapters 4 - 7 of this thesis. Individual planned objectives and contributions are as follows:

8.2.1 An improved assessment of the lahar hazards from Ruapehu volcano

Lahar deposits on the northeastern Ruapehu ring plain and along the Tongariro River were mapped and dated using interbedded tephra layers. This enabled the age, number, frequency, and distribution of Ruapehu-sourced lahars to be assessed on the northeastern Ruapehu ring plain (Chapter 5) and along the Tongariro River (Chapter 6). Dating and mapping lahar surfaces led to the development of a lahar hazard map for the entire Tongariro catchment which includes the northeastern Ruapehu ring plain (Chapter 6). Ruapehu volcano was the source of most lahars in the Tongariro catchment post- 14. 7 ka, and the source of all Holocene lahars. Eight lahar hazard zones were delineated in the Tongariro catchment ranging from 1 in > 15 000 to 1 in 35 years. Ruapehu sourced lahars occur in every hazard zone and Ruapehu is the sole source of lahars in the youngest six zones, ranging between 1 in 10 000 to 1 in 35 years. The Mangatoetoenui catchment has the greatest lahar hazard on the northeastern Ruapehu ring plain. The Upper Waikato Stream also has a high potential lahar hazard if capture of the Whangaehu River were to occur.

8.2.2 An assessment of the lahar hazards on the eastern Tongariro ring plain and the Tongariro River

The number, age, frequency and distribution of the lahars on the Tongariro ring plain were assessed using the same mapping methods as on the Ruapehu ring plain and along the Tongariro River (Chapter 5). A lahar hazard map for the Tongariro ring plain is included in the map produced for the entire catchment (Chapter 6). The eastern Tongariro ring plain is almost entirely within a 1 in > 15 000 year lahar hazard zone. Small areas of the ring plain, surrounding the Waihohonu, Oturere and Makahikatoa Streams have a greater hazard of 1 in 12 000 years. Consequently, Tongariro-sourced lahars contribute only to the two hazard zones with the longest return periods of the eight zones mapped along the Tongariro River.

8.2.3 A betterunder standing of the eruptive history (p articularly that of tephra eruptives) of the two volcanoes

This study was the first to examine sequences older than the 22.6 ka B.P. Kawakawa Tephra in the area. A tephra eruptive history of both volcanoes (but mostly Ruapehu)

176 further back in time has therefore been obtained (Chapter 3; Section 5.13; Table 5.3; Section 7.4, Fig. 7.6). From c. 75 - 64 ka, there was 1 large-scale (thick pumice lapilli-producing) eruption per 600 years on average from Ruapehu, but in the period immediately prior to 64 ka the rate was much higher. Probable Tongariro-sourced tephras were limited to the earliest part of this time interval. In the interval 64 - 36 ka there was a very low frequency of large tephra eruptions from Ruapehu, averaging 1 per 7 000 years. Between 36 and c. 23 ka the average rate of large Ruapehu eruptions was again higher, at 1 per 800 years. The rate was much higher early in the interval and very low in the latter stages. From c. 23 to 22.6 ka the rate of large Ruapehu eruptions was extremely high at 1 per 100 years. This eruption rate was equivalent or higher than that of the Bullot Formation (described by Donoghue et al., 1995), and probably indicates the start of that eruptive period. Throughout the entire 75-22.6 ka period small scale eruptions were occurring frequently, probably from both volcanoes. These are recorded by accumulations of fine ash with soil development reflecting the slow rate of surface accretion.

8.2.4 An overall synthesis of the landscape development of the northeastern Ruapehu and eastern Tongariro ring plains

Using the ring plain stratigraphy developed to meet the previous objectives, ring plain construction was examined over time and in the context of the late Quaternary geologic history of the lower North Island. This resulted in the development of a construction history of the northeastern Ruapehu ring plain from c. 75 - 22.6 ka (Section 7.3), and for both ring plains from 22.6 ka B.P. to the present. (Chapter 5). Landscape stability and paleosol development in the c.75 - 22.6 ka sequence was also elucidated (Chapter 4). A summary geologic synthesis (Section 7.4) combines all of the above information as well as additional data from geological mapping (Section 7.2).

8.3 Additional contributions made by this study

During the process of meeting the prime objectives of this study, other significant developments were made in the following areas:

8.3.1 Rhyolitic tephra identification

In previous studies, rhyolitic tephras have been identified using mineralogy (e.g. Lowe, 1988), titanomagnetite chemistry (e.g. Kahn, 1970) and glass chemistry (e.g. Froggatt, 1983). Electron microprobe-determined glass chemistry has proved to be the most successful tool, particularly when mineralogy and stratigraphic data are lacking.

177 Comparison and discrimination of glass chemistry was initially by graphical and numeric means (e.g. Borchardt et al., 1971; Froggatt, 1983). Later studies demonstrated the potential for using canonical discriminant function analysis (DFA) to discriminate glass chemistry from different sources and different tephras (e.g. Stokes and Lowe, 1988; Stokes et al., 1992; Shane and Froggatt, 1994 ). However, although these methods were used to discriminate tephras, they were not proved nor developed in a practical study in which the aim was to identify tephras. In Chapter 2 a DFA model was developed for potential tephra correlatives in the area (the point where previous studies had reached). Following on from this, unknown tephras were identified using probabilities of their classification to tephras in the model. This was the first practical demonstration of these techniques on New Zealand tephras. In addition, methods of improving the identification of the tephras were tried, including addition of correctly classified previously unknown tephras to the original model, and treatment of some unknowns as mixed tephras.

8.3.2 Andesitic tephra discrimination

The DFA methods used for rhyolitic tephra discrimination had never been used for andesitic tephras in New Zealand. In the first part of Chapter 3 (Section 3.3) the DFA methods were tested on electron microprobe-determined chemistry of various phenocryst phases to discriminate between two volcanic sources (Ruapehu and Egmont volcanoes) and then between individual tephras of one of the sources (Egmont). This study established that these discriminations were possible and that titanomagnetite was the best mineral phase for this purpose. In the second part of Chapter 3 (Section 3.4) the use of DFA and clustering techniques were demonstrated for development of an andesitic tephrostratigraphy in a previously unknown sequence. These techniques were shown to be able to distinguish marker tephras which could be used in correlation.

8.3.3 Hal/oysite/allophane fo rmation in andesitic tephras

In Chapter 4 (Section 4.9), a two stage weathering process was shown to account for an apparently unusual combination of 2:1 allophane and halloysite present in a paleosol sequence developed into andesitic tephras. Most studies of weathering tephras had not taken such a process into serious consideration. lt is commonly considered (e.g. Parfitt et al. , 1984; Singleton et a/., 1989) that soil solution Si/ AI ratios determine whether 2:1 allophane, 1:1 allophane or halloysite is formed from weathering tephras. Soil solutions high in Si promote formation of 1:1 allophane and halloysite, low Si:AI ratios promote 2:1 allophane. In andesitic tephras, lower Si:AI ratios normally promote the formation of 2:1 allophane (e.g. Kirkman 1981; Lowe, 1986). This occurred in the Ruapehu tephra sequence studied when the tephra/soil surface was originally accreting. Later rapid burial of this weathered material

178 by 20 m of lahar deposits and tephras created poor drainage conditions within it. In addition, weathering of the overburden resulted in large amounts of dissolved silica leaching downward into the buried tephra. This resulted in ongoing weathering of remaining andesitic glass into halloysite rather than the 2:1 allophane previously formed at the soil surface.

8.3.4 Relationship of ring-plain construction to late Quaternaryclim ate change

A well constrained rhyolitic tephrochronology was developed for the Ruapehu and Tongariro ring-plain sequences in Chapter 2. This enabled examination of the ring plains' depositional history in relation to other records of late Quaternary climate change in the lower North Island. In Section 4.1 0, periods of paleosol development in the Ruapehu ring-plain sequence were correlated to periods of relatively warm climate and widespread soil development in the lower North Island. In Section 7.3 and 5.13, periods of large scale and widespread lahar and streamflow deposition on the ring plains were correlated to the last and antepenultimate stadials of the Last Glacial (Ohakean and Porewan stadials). The Ruapehu and Tongariro ring plains therefore record a composite history of both volcanic events and late Quaternary climate change.

8.4 Potential future work

A) An examination of the so-called middle Tongariro Subgroup tephras (Milne and Smalley, 1979) preserved in loess sections to the south, east and west of the study area. Correlation of these to their near source equivalents would prove valuable in providing further chronology for the loess sequences. In addition, the provenance and correlation of tephras in loess sequences to the west of Ruapehu (which could contain tephras from both Egmont and the Tongariro Volcanic Centre) could be established.

B) A pedological and mineralogical investigation of the Papakai Formation around the entire Tongariro Volcanic Centre. This study would provide valuable information on soil-forming and weathering processes within andesitic tephra of the same age in areas of different drainage and climate. lt may also provide additional information about the climate in this area during the Holocene.

C) A detailed investigation of the upper ring-plain portions of the Upper Waikato Stream and the Whangaehu River. The purpose of this study would be to (1) identify definitely if lahars from the Whangaehu River have entered the Upper Waikato Stream, and (2) examine the potential for this to happen in the future, following the events of 1995.

D) Investigate the physical volcanology of the Mangamate Tephra, Bullet Formation and older tephras. This would provide information on column height, eruption rates and

179 potential distributions of these larger events from Ruapehu and Tongariro. Information such as this could be used to predict and prepare for the effects of larger eruptions of either of the two volcanoes in future.

E) Investigation of the large lava flow(s) along the Tongariro River. This unit has not been adequately dated, nor has its origin been completely determined. Is it a series of large flows derived from Tongariro volcano or are the flows from satellite vents or a fissure along the eastern TVZ boundary fault close to the present course of the Tongariro River?

8.5 References

Borchardt, G.A..; Harward, M.E.; Schmitt, R.A., 1971 . Correlation of ash deposits by activation analysis of glass separates. Quat. Res. 1: 247-260. Donoghue, S.L.; Neall, V.E.; Palmer, A.S., 1995. Stratigraphy and chronology of late Quaternary andesitic tephra deposits, Tongariro Volcanic Centre, New Zealand. J. Roy. Soc. N. Z. 25: 115-206. Froggatt, P.C., 1983. Towards a comprehensive Upper Quaternaryte phra and ignimbrite stratigraphy in New Zealand using electron microprobe analysis of glass shards. Quat. Res. 19: 188-200. Kirkman, J.H., 1981 . Morphology and structure of halloysite in New Zealand tephras. Clays and Clay Min., 29, 1-9. Kohn, B.P. , 1970. Identification of New Zealand tephra layers by emission spectrographic analysis of their titanomagnetites. Lithos 3: 361-368. Lowe, D.J., 1986. Controls on the rates of weathering and clay mineral genesis in airfall tephras: a review and New Zealand case study. In: Colman, S.M., and Dethier, D.P. (eds) Rates of chemical weathering of rocks and minerals. Academic Press, Orlando, pp. 265-330. Lowe, D.J., 1988. Stratigraphy, age, composition, and correlation of late Quaternary tephras interbedded with organic sediments in Waikato lakes, North Island, New Zealand. N. Z. J. of Geol. and Geophys. 31: 125-165. Milne, J.D.G. and Smalley, I.J., 1979. Loess deposits in the southern part of the North Island of New Zealand: an outline stratigraphy. Acta Geologica Academiae Scientarium Hungaricae, 22: 197-204. Parfitt, R.L., Saigusa, M. and Cowie, J.D., 1984. Allophane and halloysite formation in a volcanic ash bed under different moisture conditions. Soil Sci., 138, 360-364. Shane, P.A.R., Froggatt, P.C., 1994. Discriminant function analysis of glass chemistry of New Zealand and North American tephra deposits. Quat. Res. 41: 70-81. Singleton, P.L., Mcleod, M. and Percival, H.J., 1989. Allophane and halloysite content and soil solution silicon in soils from rhyolitic volcanic material, New Zealand. Australian J. of Soil Res., 27, 67-77.

180 Stokes, S.: Lowe, D.J., 1988. Discriminant function analysis of late Quaternary tephras from five volcanoes in New Zealand using glass shard major element chemistry. Quat. Res. 30: 270-283. Stokes, S.; Lowe, D.J.; Froggatt, P.C., 1992. Discriminant function analysis and correlation of Late Quaternary rhyolitic tephra deposits from Taupo and Okataina volcanoes, New Zealand, using glass shard major element composition. Quat. lnter. 13/14: 103-1 17. Topping, W.W., 1973. Tephrostratigraphy and chronology of late Quaternary eruptives from the Tongariro Volcanic Centre, New Zealand. N. Z. J. Geol. and Geophys., 16: 397-423. Topping, W.W., Kohn, B.P., 1973. Rhyolitic tephra marker beds in the Tongariro area, North Island, New Zealand. N.Z. J. Geol. and Geophys., 16: 375-395.

181 APPENDIX 1: SAS PROGRAMS USED IN THIS STUDY

A 1.1 Programs

The analyses within Chapter 2 were all carried out using SAS Release 6.11 for Windows 95. Chapter 3 analyses were carried out using SAS Release 5.0 on a UNIX system.

A) Glass data were entered from tab-delimited text files, constructed in the following format: Sample 11 Group // Si02 // Ti02 I/ Al203 // FeO // CaO I/ Na20 I/ K20. Consequently, the Data Step required for all of the programs was: data libname.sas-dataset-name; infile "absolute address of text file"; input sample group sio2 tio2 al2o3 feo cao na2o k2o; run;

The sample field was a number individually identifying each analysis of each sample. The group field was an integer code, each number represented a different tephra. The other fields contained the log-transformed oxide values.

B) To construct discriminant models the following program step was used: proc discrim data=libname.sas-dataset-name outstat=libname.outstat-dataset-name ncan=5 listerr distance simple; class group; var sio2 tio2 al2o3 feo cao na2o k2o; id sample; run;

The "outstat" option was used to store the discriminant model parameters which were required to test the unknown samples against. The number of canonical variables was the lesser number of 1-(number of tephra groups) and 1-(number of oxide predictors). The "listerr'' option prints the observations that were classified incorrectly and lists their probabilities of classification to each of the tephra groups in the discriminant model. The 2 "distance" option prints the Mahalanobis (0 ) distances between tephra groups in the model. The "simple" option prints means and standard deviations of each of the tephra groups.

A 1 C) To produce the plots of canonical variates for the discriminant models the following program steps were used:

proc candisc data=libname.sas-dataset-name out=libname.outcan-dataset-name; class group; var sio2 tio2 al2o3 feo cao na2o k2o; run; proc plot data=libname.outcan-dataset-name; plot can2*can1 =group I href=O vref=O; run;

All canonical variables are stored in the "out" dataset by the first part of the procedure. The second part of the procedure plots the first two canonical variables from the "out" dataset.

D) To test unknown analyses against the discriminant models the following program step was used:

proc discrim data=libname.outstat-dataset-name testdata=libname.test-dataset-name testout=libname.testout-dataset-name testlist; class group; var sio2 tio2 al2o3 feo cao na2o k2o; testclass group; testid sample; run;

This program uses the "outstat" dataset created when the discriminant model was initially constructed and tests the unknown analyses (test-dataset) against it. The "testlist" option prints probabilities of classification of all of the unknown samples with each of the tephra groups in the discriminant model.

E) To establish the variables which discriminate between tephra groups the most by stepwise DFA, the following program step was used: proc stepdisc data=libname.sas-dataset-name stepwise; class group; var sio2 tio2 al2o3 feo cao na2o k2o; run;

A2 F) To carry out clustering analysis of the tephra samples the following program steps were used:

proc cluster data=libname.sas-dataset-name method=wards ccc pseudo; var sio2 tio2 al2o3 feo cao na2o k2o; copy group; proc tree ncl=(number of clusters you suspect) out=libname.outclus-dataset; copy sio2 tio2 al2o3 feo cao na2o k2o group; proc freq; tables cluster*group;

The first program step carries out the clustering process using a minimum variance method. The second step plots a tree diagram to help interpret the number of clusters present. The third step produces a table comparing the cluster groupings with those which you have pre-defined. Usually a "candisc" step was added to this procedure to evaluate the efficiency of the clustering, with the class=cluster.

Note: For each program or data step the list of variables was customised for the particular mineral or glass phase being considered.

A 1.2 Reference

SAS Institute Inc., 1989. SAS users guide: statistics. Version 6 Edition. Cary N.C. , SAS Institute Inc., 1028 pp.

A3 Appendix Rhyolitic glass 2:

Appendix 2: Rhyolitic glass analyses

All of the glass analyses carried out in this study follow. Samples are listed in the same order as Table 2.2 (Chapter 2). The analyses are in a raw form (not normalised).

Sample Location Analysis Si02 Ti02 A1203 FeO M gO CaO Na20 1<20

Sample 93.5 T20/465127 76.050 0.529 11.899 0.792 0.190 0.732 4.027 3.524 2 75.563 0.278 12.225 1.059 0.161 0.995 4.156 3.107 3 76.593 0.509 12.004 1.109 0.120 1.069 4.128 3.825 4 76.788 0.345 11.923 1.049 0.108 1.038 4.183 3.687 5 76 .175 0.522 12.126 0.897 0.065 1.153 4.237 3.826 6 78.617 0.748 12.012 0.909 0.191 1.084 4.236 3.443 7 75.637 0.316 12.064 1.032 0.121 1.048 3.896 3.783 8 76.51 1 0.340 12.005 0.992 0. 122 1.323 4.146 3.320

Mean 76.492 0.448 12.032 0.980 0.135 1.055 4.126 3.564

Sample 93.6 T20/453129 1 77.442 0.187 12.734 1.029 0.232 0.910 4.301 3.333 2 73.210 0.175 11.845 1.074 0.226 0.925 4.158 3.156 3 75.842 0.163 12.403 1.093 0.221 0.984 4.302 3.429 4 74.592 0.164 12.058 1.031 0.128 0.965 4.352 3.478 5 75.656 0.168 12.384 0.973 0.134 0.900 4.314 3.357 6 79.033 0.170 12.604 0.906 0.234 0.998 4.152 3.303 7 78.614 0.088 12.381 0.902 0.215 0.834 3.876 3.386 8 77.977 0.205 12.357 1.010 0.161 0.962 4.225 3.359 9 76 .916 0.229 12.157 1.124 0. 141 0.925 4.147 3. 135 10 79.594 0.222 11.887 0.926 0.197 0.887 4.346 3.428 Mean 76.888 0.177 12.281 1.007 0.189 0.929 4.21 7 3.336

Sample 93.7 T20/430127 77.655 0.694 12.323 0.830 0.174 0.957 3.956 3.977 2 74.621 0.232 11.480 0.982 0.191 1.045 4.190 3.417 3 76.167 0.599 11.639 0.718 0.187 0.931 3.995 3.087 4 75.672 0.319 12.235 0.918 0.124 0.909 4.149 3.037 5 75.572 0.269 11.826 0.763 0.212 0.728 4. 175 3.226 6 75.186 0.414 11.945 0.889 0.072 0.815 4.035 2.949 7 74.603 0.209 12.058 0.783 0.231 0.991 3.942 3.066 8 74. 134 0.217 12.338 0.855 0.144 0.708 3.713 3.138 9 75.609 0.208 12.017 1.155 0.154 0.894 4.056 3.393 Mean 75.469 0.351 11.985 0.877 0.165 0.886 4.023 3.254

Sample 93.8 T20/435129 1 76.975 0.557 12.197 0.832 0.186 1.350 4.403 3.687 2 77.440 0.624 11.961 1.093 0.136 1.613 4.079 3.379 3 71 .81 2 0.861 11.792 0.935 0. 122 1.947 3.962 3.359 4 77.496 0.243 12.469 1.260 0. 156 0.828 4.240 3.684 5 74.878 1.094 12.002 1.217 0. 152 1.071 3.925 3.684 6 77.166 0.000 12.542 0.989 0. 180 1.678 4.123 3.316 7 77.280 0.074 12.337 0.976 0.235 1.331 4.105 3.682 8 74.892 0.491 11.523 1.177 0.172 1.127 4.072 3.210 9 78.737 0.866 11.978 1.082 0.163 0.814 4.144 3.625 10 73.917 1.471 11.972 0.980 0.177 0.384 3.956 3.578 11 77.022 1.320 11.678 1.111 0.101 1.026 4.171 3.958 12 78.007 0.926 12.231 0.933 0. 183 0.952 4.102 3.778 13 77.894 0.301 11.901 1.021 0. 114 1.935 4.069 3.235 14 75.064 0.936 11.304 1.001 0.145 1.242 3.902 3.329 Mixed

Sample 93.9 T20/498157 74.197 0.106 12.017 0.987 0.074 0.755 4.220 3.457 2 75.680 0.1 10 12.022 0.964 0.065 0.535 4. 135 3.721 3 73.492 0.167 12.266 1.058 0.202 0.877 4.420 3.499 4 74.504 0.075 11.851 0.945 0.084 0.536 4. 164 3.776 5 74.445 0.130 11.949 1.168 0.118 0.725 4.451 3.305 6 73.919 0.113 11.804 1.192 0.166 0.601 4.323 3.296 7 73.278 0.125 11.950 1.066 0.132 0.577 4.023 3.31 1 8 76. 138 0.106 12.164 0.913 0.074 0.561 4.200 3.745 9 75.713 0.179 11.949 1.080 0.128 0.655 4.278 3.458 10 73.635 0.120 12.255 1.337 0.186 0.856 4.187 3.250 11 73. 199 0.266 11.818 1.131 0.147 0.764 4.288 3.831 Mean 74.382 0.136 12.004 1.076 0.125 0.677 4.244 3.514

A4 Appendix 2: Rhyolitic glass

Sample 93.32 T20/469100 1 74.862 0.147 11.925 0.868 0.052 0.658 3.791 4.428 2 74.338 0.131 12.169 0.844 0.067 0.764 3.950 4.266 3 74.872 O.D78 12.390 0.954 0.109 0.698 3.979 4.347 4 75.023 0.131 12.363 0.882 0.048 0.717 4.059 4.191 5 74.542 0.090 12.069 0.870 0.064 0.554 3.792 4.408 6 71.756 0.122 12.010 0.838 0.072 0.634 3.766 4.161 Mean 74.232 0. 117 12.154 0.876 0.069 0.671 3.890 4.300

Sample 93.59 T19/505248 78.01 1 0.229 13.106 1.280 0.257 1.559 4. 150 2.814 2 74.733 0. 196 12.353 0.773 0. 129 0.717 3.707 4.056 3 74.021 0. 125 12.101 1.111 0.197 0.869 4.291 3.421 4 78.826 0.272 13.324 1.336 0.278 1.370 4.091 3.528 5 75.769 0.246 12.403 0.752 0.063 0.638 3.744 4.620 6 75.989 0. 136 12.285 0.780 0.045 0.738 3.911 4. 186 7 76.539 0. 129 12.338 0.928 0.082 0.523 3.993 4.375 8 76.667 0. 128 11.801 0.996 0.070 0.649 4.041 4.031 9 75.320 0.273 12.318 0.981 0.109 0.991 4.1 17 3.685 10 78.489 0.159 12.307 1.075 0.156 0.933 4.124 3.455 Mixed

Sample 94.7 T20/465127 1 75.627 0.1 14 12.012 1.017 0.162 0.898 3.874 3.310 2 76 .161 0.104 12.152 0.988 0.143 0.834 3.749 3.200 3 72.616 0.101 11.572 2.004 0.192 0.829 3.756 3. 154 4 75.023 0.260 11.801 0.883 0.168 0.871 3.946 3.087 5 76.403 0. 186 12.429 0.834 0.150 0.928 3.858 3.263 6 72.965 0. 147 11.638 0.737 0.083 0.839 3.632 2.901 7 76.166 0. 163 12.166 1.052 0.175 0.913 3.848 3.132 8 76.104 0. 176 12.235 0.972 0.141 1.121 3.571 3.398 9 74.466 0. 155 11.854 0.972 0.1 15 0.965 4.003 3. 118 Mean 75.059 0. 156 11.984 1.051 0. 148 0.91 1 3.804 3. 174

Sample 94.9 T20/468 162 76.328 0.214 12.883 1.740 0.198 1.380 3.912 3.028 2 74.926 0.184 12.729 1.530 0.174 1.418 3.942 2.925 3 72.836 0.148 11.768 1.315 0.134 1.043 3.604 2.904 4 75.351 0.237 12.771 1.528 0.147 1.220 3.933 2.854 5 75.859 0.190 12.776 1.506 0.200 1.454 3.938 2.893 6 75.876 0.181 12.891 1.616 0.179 1.396 3.885 3.138 7 74.500 0.189 12.802 1.548 0.165 1.410 4.026 2.895 8 72.239 0.244 11.820 1.492 0.189 1.272 3.931 2.981 9 76.376 0.179 12.921 1.577 0.194 1.519 4.207 3.172 10 72.517 0.308 12.559 1.321 0.255 1.365 4.236 2.717 Mean 74.681 0.207 12.592 1.517 0.184 1.348 3.961 2.951

Sample 94.21 T19/48121 1 1 73.883 0.206 11.760 1.015 0.185 1.013 3.752 3.270 2 73.903 0.207 12.042 0.998 0.237 0.854 3.944 3.029 3 75.398 0.198 12.014 0.922 0.161 0.863 3.781 3.482 4 75.472 0.175 12.628 0.972 0.187 1.274 4.144 3.119 5 76.873 0.157 11.933 0.879 0.197 0.888 3.939 3.282 6 75.862 0.216 11.066 1.147 0.121 0.663 3.531 3.364 7 72.994 0.093 11.640 0.871 0.209 0.927 3.791 2.907 8 74.943 0.198 12.201 1.171 0.195 0.888 3.938 3.167 9 74.222 0.095 12.026 0.964 0.145 0.787 3.797 3.583 Mean 74.839 0.172 11.923 0.993 0.182 0.906 3.846 3.245

Sample 94.24 T19/48621 1 1 75.850 0.179 12.800 1.456 0. 167 1.531 4.065 2.920 2 75.608 0.256 12.941 1.766 0.172 1.433 4.022 2.909 3 75.050 0.187 12.873 1.570 0.162 1.546 3.984 3. 188 4 72.989 0.269 12.590 1.448 0.240 1.498 3.960 3.022 5 74.509 0.215 12.765 1.543 0.246 1.651 3.884 3. 123 6 74.391 0.242 12.786 1.501 0.209 1.484 4039 3.075 7 75.569 0.214 12.877 1.762 0. 156 1.402 3.936 2.867 8 75.009 0.243 12.686 1.786 0.161 1.262 4.220 2.986 9 71 .286 0.171 16.033 1.339 0. 144 2.982 4.779 2.329 10 72.349 0.258 12.495 1.604 0.168 1.332 4.058 2.945 Mean 74.261 0.223 13.085 1.578 0.183 1.612 4.095 2.936

Sample 94.27 T19/485221 79.743 0.057 12.912 1.125 0.078 0.702 3.679 3.991 2 74.259 0.155 11.758 1.002 0.050 0.724 3.602 4.070 3 75.002 0.214 11.848 0.959 0.134 0.851 3.901 3.109 4 75.289 0.158 11.792 1.097 0.125 1.053 3.883 3.144 5 77.128 0.109 12.314 0.882 0.096 0.795 3.775 3.958 6 77.463 0.149 12.317 1.139 0.202 0.838 4.013 3.056 7 75.757 0.222 11.997 1.067 0.158 0.834 4.002 3.193

A5 Appendix 2: Rhyolitic glass

8 75.999 0. 103 12.190 0.868 0.071 0.833 3.865 3.387 9 76.081 0. 158 12.170 1.039 0.101 0.982 4.017 3.299 Mean 76.302 0. 147 12. 144 1.020 0.113 0.846 3.860 3.467

Sample 94.28 T1 9/495220 73.280 0.052 11.670 1.174 0.080 0.755 3.381 3.166 2 74.430 0.061 11.824 0.941 0.131 0.755 3.660 3.792 3 71 .086 0. 106 11.596 1.357 0.072 0.828 3.796 3.041 4 73.820 0. 103 11.752 0.904 0. 126 0.702 3.128 4.248 5 74.483 0. 115 11.754 1.197 0.121 0.778 3.737 3.956 6 70.450 0. 102 11.309 1.111 0.105 0.972 3.551 3.016 7 73.917 0. 128 11.680 1.067 0. 106 0.664 3.761 3.525 8 72.529 0.070 11.663 0.906 0.1 14 0.791 3.715 3.602 9 71 .928 0. 128 11.423 1.199 0.167 0.573 3.652 3.443 10 70.504 0. 164 11.583 1.257 0.145 0.982 3.637 3.108 Mixed

Sample 94.29 T19/495220 1 73.167 0.073 11.836 1.089 0.123 0.709 2.672 3.505 2 72.856 0. 157 11.542 1.190 0.045 0.759 3.107 3.575 3 72.770 0.105 11.552 1.003 0.043 0.732 3.282 3.774 4 73.401 0.219 12.000 1.31 1 0.106 0.944 3.488 3.316 5 73.704 0. 174 12.088 1.374 0.164 0.871 3.517 3.424 Mixed

Sample 94.37 T1 9/485238 76.275 0.147 12.248 1.058 0.1 16 0.929 4.022 3.319 2 73.883 0.090 11.687 1.055 0.178 0.856 3.969 2.951 3 73.701 0.181 11.499 1.113 0.171 0.880 3.777 2.834 4 78.023 0.107 12.327 1.127 0.140 0.897 4.095 3.347 5 76.21 7 0.157 12.014 0.928 0.174 0.958 4.208 3. 145 6 76.923 0.158 12.41 1 1.137 0.184 0.867 3.890 3.252 7 76.462 0.105 12.176 1.155 0.242 0.965 3.767 3.104 8 74.827 0. 144 12.423 1.040 0.170 1.092 3.874 3.038 9 77.566 0.1 13 12.040 0.904 0.146 0.795 3.859 3.280 10 74.951 0.158 11.833 0.961 0.183 0.931 3.941 3.107 Mean 75.883 0. 136 12.066 1.048 0.170 0.917 3.940 3.138

Sample 94.47 T1 9/496238 1 76.533 0. 120 11.293 1.270 0.055 0.636 3.491 3.813 2 76.019 0. 105 11.427 0.922 0.003 0.607 3.777 3.707 3 75.026 0. 169 11.337 0.964 0.043 0.563 3.665 3.718 4 75.729 0.151 11.268 0.975 0.044 0.510 3.647 3.972 5 75.787 0. 145 11.389 1.058 0.078 0.489 3.649 3.496 6 76 .166 0.175 11.368 1.043 0.094 0.390 3.713 3.775 7 75.920 0.175 11.418 0.863 0.034 0.360 3.664 3.678 8 74.862 0.137 11.123 1.006 0.074 0.377 3.461 3.753 9 75.598 0.251 11.412 0.961 0.098 0.396 3.845 3.538 10 75.021 0.122 11.232 0.988 0.137 0.393 3.528 3.727 Mean 75.666 0.155 11.327 1.005 0.066 0.472 3.644 3.718

Sample 94.51 T1 9/504237 1 76.588 0.178 12.134 0.799 0.148 0.871 3.414 2.963 2 73.653 0.203 11.639 1.008 0. 198 0.842 3.662 3.246 3 75.575 0.179 11.992 0.753 0.140 0.916 3.791 3.168 4 76.264 0.176 12.149 0.892 0.168 0.795 3.578 3.417 5 75.477 0.122 12.090 0.860 0. 138 0.919 3.909 3.019 6 73.925 0. 186 11.759 0.835 0. 112 0.820 3.811 3.016 7 72.491 0.180 11.540 1.058 0. 172 0.950 3.669 2.868 8 72.490 0. 177 11.518 1.104 0.1 12 0.888 3.788 3.002 9 70.81 1 0. 124 11.312 1.060 0.103 0.716 3.572 2.803 Mean 74.142 0. 169 11.793 0.930 0.143 0.857 3.688 3.056

Sample 94.58 T19/505248 1 75.099 0.054 12.014 0.759 0.098 0.856 3.962 3.152 2 75.708 0.128 11.926 0.887 0.189 0.961 4.084 3.215 3 73.637 0.1 19 11.646 0.637 0.119 0.832 3.734 3.099 4 75.082 0.126 11.821 0.733 0. 126 0.768 3.640 3.747 5 73.992 0.148 11.869 0.570 0.121 1.094 4. 108 2.943 6 72.901 0.096 11.766 1.098 0.274 1.062 3.623 3.182 7 76.702 0. 192 12.349 0.792 0.1 11 0.941 3.932 3.330 8 72.265 0.077 11.496 0.783 0. 193 0.886 3.476 3.002 9 72.113 0.077 12.394 0.537 0.093 0.871 4.323 2.590 Mean 74.167 0.1 13 11.920 0.755 0.147 0.919 3.876 3.140

Sample 94.71 T1 9/488242 1 75.523 0.036 12.082 0.799 0.142 0.854 3.51 1 4.088 2 73.692 0.173 12.096 0.819 0.134 0.791 3.620 4. 152 3 76.017 0.189 12.372 0.906 0.171 0.979 4.054 3.284 4 76.783 0.115 12.066 1.063 0.148 0.892 4. 149 3.048 5 74.283 0.105 11.962 0.981 0.170 0.944 3.854 3.327

A6 Appendix 2: Rhyolitic glass

6 76.948 0.130 11.994 1.039 0.162 0.970 3.853 3.447 7 74.888 0.136 12.089 0.748 0.164 0.909 3.873 3.337 8 75.921 0.173 12.439 1.102 0.086 0.935 3.919 3.888 9 74.191 0.060 12.172 0.971 0.123 0.688 3.844 3.860 10 73.835 0.093 12.198 0.923 0.090 1.000 3.957 4.090 Mixed

Sample 94.74 T1 9/499247 1 77.275 0. 134 12.184 1.110 0. 174 0.880 3.980 3.133 2 76.234 0.104 11.997 1.248 0. 129 0.921 3.804 3.137 3 74.633 0.108 11.938 0.975 0.161 0.764 4.425 3.099 4 75.454 0. 160 12.015 0.996 0.146 0.818 3.891 3. 174 5 76.484 0.307 12.061 0.912 0.210 0.877 3.777 3. 178 6 74.212 0.225 11.823 1.076 0.186 0.831 3.902 2.904 7 75.794 0.336 12.131 0.935 0.177 0.823 3.897 3.062 8 75.541 0.115 12.137 0.884 0.110 0.867 3.990 3.161 9 74.963 0.107 12.323 1.045 0.131 0.81 1 4.173 3.239 10 75.025 0.130 12.053 1.084 0.162 0.829 4.022 3.324 Mean 75.562 0.173 12.066 1.027 0.159 0.842 3.986 3.141

Sample 94.79 T19/364544 1 75.220 0.338 13.638 1.889 0.255 1.441 3.849 2.987 2 74.605 0.405 13.438 2.140 0.335 1.461 4.057 3.013 3 74.716 0.285 13.377 1.992 0.303 1.388 4.150 2.891 4 75.099 0.384 13.207 1.976 0.359 1.402 4. 136 2.938 5 75.133 0.254 13.345 1.897 0.320 1.412 4.036 2.998 6 74.998 0.345 13.002 1.995 0.338 1.442 4.1 12 2.901 Mean 74.962 0.335 13.335 1.982 0.318 1.424 4.057 2.955

Sample 95.2 T1 9/430238 1 75.323 0.201 11.770 0.844 0. 155 0.896 3.878 2.920 2 76.403 0.221 12.033 0.939 0.191 0.894 3.848 3.221 3 74.617 0.145 11.975 0.886 0.134 0.871 4.193 2.934 4 74.023 0.173 11.920 1.002 0.116 0.898 3.916 3.008 5 76.586 0.130 12.206 1.177 0.186 0.914 3.668 3.772 6 71.486 0.183 11.352 0.891 0.123 0.827 3.670 2.825 7 75.139 0. 184 11.883 1.125 0.204 0.883 3.958 3.231 8 76.564 0. 131 12.139 1.013 0.151 0.869 2.956 3.488 9 76.362 0. 182 12.329 0.997 0.121 0.898 3.991 3.073 Mean 75. 167 0. 172 11.956 0.986 0. 153 0.883 3.786 3.164

Sample 95.3 T1 9/430238 1 74.844 0. 112 12.187 1.084 0.059 0.641 3.538 4. 144 2 73.657 0.063 11.775 0.867 0.033 0.581 3.473 4.1 16 3 74.676 0.080 11.900 0.873 0.078 0.671 3.549 3.898 4 73.806 0. 165 11.796 0.907 0.058 0.603 3.621 4.039 5 75.048 0.094 12.345 1.006 0.060 0.779 3.718 3.860 6 72.598 0. 109 11.932 1.004 0.084 0.682 3.542 3.793 7 74.831 0. 169 12.290 0.861 0. 135 1.014 3.971 2.883 8 75.977 0.296 12.212 1.192 0.241 1.196 3.854 3. 160 Mean 74.430 0.136 12.055 0.974 0.094 0.771 3.658 3.737

Sample 95.6 T19423339 75.755 0.114 12.105 1.184 0.220 0.956 3.889 3.163 2 75.375 0.133 11.969 1.038 0.233 0.905 3.940 3.285 3 73.722 0.201 11.846 0.987 0.124 0.862 3.869 3.099 4 75.129 0.165 11.752 1.056 0.206 0.980 4.070 3.162 5 73.909 0.123 11.647 1.096 0.124 0.921 3.783 3.253 6 76.427 0.190 12.141 1.062 0.236 1.042 4.044 3. 174 7 74.366 0.206 12.156 1.020 0.190 0.985 3.828 3.218 8 74.891 0.161 11.854 1.098 0.144 0.834 3.902 3. 160 9 73.660 0.142 11.989 1.069 0.205 0.938 3.837 3.051 10 75.950 0.110 11.935 1.118 0. 151 0.941 3.861 3.255 Mean 74.918 0.155 11.939 1.073 0. 183 0.936 3.902 3. 182

Sample 95.9 T1 9/448326 1 74.315 0.261 11.550 0.857 0. 154 0.852 4.003 3.049 2 73.074 0.205 11.616 0.844 0.1 17 0.910 3.848 2.841 3 74.463 0.240 11.602 1.025 0.252 0.874 3.784 3.035 4 75.748 0. 179 12.264 0.928 0.158 0.782 3.908 3.577 5 73.373 0.094 11.978 1.068 0.140 0.851 3.677 2.909 6 76.560 0.212 12.346 1.070 0.165 0.867 3.969 3.348 7 76.747 0.146 12.178 0.924 0.195 0.904 3.825 3.236 8 76.238 0. 179 12.245 0.964 0.177 0.925 4.300 3.114 9 74.495 0.246 12.009 1.120 0.204 0.958 4.053 3.174 10 72.904 0.167 11.770 1.080 0.176 0.958 3.848 2.963 Mean 74.792 0.193 11.956 0.988 0.174 0.888 3.922 3.125

Sample 95. 13 T1 9/549328 79.480 0. 119 12.638 1.026 0.106 1.032 3.710 3.034 2 74.747 0.164 11.955 0.979 0.131 0.928 3.834 2.860

A? Appendix 2: Rhyolitic glass

3 75.815 0.114 12.027 0.898 0.154 0.893 3.825 3.050 4 71.140 0.123 11.304 0.881 0.122 0.862 3.660 2.757 5 71 .762 0.242 11.439 1.018 0. 198 0.918 3.558 2.836 6 74.703 0.182 11.812 0.983 0.239 0.789 3.798 3.161 7 73.479 0.130 11.549 1.016 0.145 0.905 3.813 2.935 8 75.169 0.178 11.705 1.091 0.191 0.878 3.936 3.21 1 9 71 .442 0.213 11.494 1.041 0.196 0.904 3.656 2.723 Mean 74. 193 0.163 11.769 0.993 0.165 0.901 3.754 2.952

Sample 93.1 T20/455088 1 74.231 0.670 11.539 1.210 0.142 0.926 3.954 3.422 2 73. 136 0.384 11.953 1.233 0.121 1.285 3.949 3.332 3 72.679 0.563 11.949 1.249 0.121 1.287 3.945 3.247 4 73.000 0.224 11.425 1.029 0.248 1.141 3.800 3. 169 5 73.659 0.354 11.741 1.278 0.068 1.229 3.941 3.356 6 75.709 0.329 11.653 1.162 0.130 1.595 3.968 3. 189 7 75.540 0.261 11.655 1.234 0.109 1.455 4.051 3.243 8 76 .139 0.833 11.453 1.224 0. 106 1.264 3.577 3.332 9 73.218 0.297 11.828 1.290 0.153 1.603 4.108 2.996 10 75.809 0.653 11.650 1.099 0.083 1.283 3.969 3.403 Mean 74.312 0.459 11.685 1.201 0. 128 1.307 3.926 3.269

Sample 93.3 T20/465095 1 70.768 0. 173 11.753 1.175 0.185 0.949 4.003 3.159 2 70.197 0.191 11.648 1.286 0. 116 1.160 4.030 2.882 3 73.239 0. 132 12.168 1.258 0. 158 1.052 4.071 3.202 4 73.307 0.055 11.634 1.249 0.154 0.970 4.138 3.249 5 74.545 0.148 11.684 1.374 0. 109 0.983 3.979 3.141 6 72.958 0.144 11.979 1.328 0. 157 1.127 4.247 3. 196 7 72.515 0.325 12.145 1.133 0.114 0.987 4. 263 3.302 8 73.907 0.200 12.197 1.226 0.164 1.064 4.250 3. 153 9 75.796 0. 125 11.757 1.240 0.143 1.068 4.016 3.228 10 75.877 0.098 11.365 1.050 0.076 1.021 3.841 3.230 Mean 73.31 1 0.159 11.833 1.232 0.138 1.038 4.084 3.174

Sample 93.17 T20/46 1170 72.683 0.205 12.208 1.259 0.156 1.148 3.938 2.798 2 75.642 0.120 12.252 1.071 0.139 0.936 4.184 3.268 3 75.094 0.146 12.090 1.156 0.142 1.134 4.304 2.963 4 74.458 0.136 11.842 0.950 0.095 0.693 3.746 3.993 5 73.607 0. 181 12.368 0.998 0.195 1.070 4.357 3.007 6 73.175 0.236 11.785 1.424 0.181 1.057 4.026 2.791 7 73.396 0. 115 12.002 1.142 0. 151 0.975 4.103 2.999 8 74.019 0.225 12.098 1.435 0.225 1.034 4.236 3.003 9 72.013 0.141 11.904 1.465 0.147 1.060 3.877 2.820 10 73.893 0.137 12.258 1.455 0.190 1.038 4.281 3.154 11 74.964 0.263 12.149 1.400 0.151 1.130 4.140 2.955 Mean 73.904 0.173 12.087 1.250 0.161 1.025 4.108 3.068

Sample 93.33 T20/469100 72.696 0.276 11.862 1.374 0.244 1.514 3.577 3.285 2 72.563 0.221 11.229 1.164 0.150 1.032 3.912 3.340 4 75. 141 0.154 12 016 1.164 0.088 1.107 4.018 2.986 5 71 .388 0.321 11.598 1.320 0.142 1.186 4. 125 2.990 6 77.992 0.200 11.974 1.299 0.154 1.519 4.132 3.597 7 77.409 0.143 11.733 1.218 0.122 1.093 4. 126 3.412 8 75.353 0.236 11.997 1.078 0.116 0.91 1 3.750 4.313 9 70.170 0.274 11.658 1.156 0.156 1.076 3.741 2.983 10 73.820 0.083 12.594 1.434 0.175 1.237 4.206 3.067 11 76.510 0.142 12.510 1.322 0.144 1.357 4. 182 3.466 Mean 74.304 0.205 11.917 1.253 0.149 1.203 3.977 3.344

Sample 7.37 T20/469100 73.665 0.158 12.710 1.206 0.099 1.507 3.385 3.547 2 73.105 0.268 12.042 1.824 0.120 1.317 2.832 3.509 3 74. 123 0.156 12.265 1.645 0.142 1.332 3.698 3.269 4 73. 125 0.251 12.740 1.468 0.125 1.471 3.889 3.358 5 73.235 0.136 12.059 1.338 0.145 1.358 3.678 3.332 6 73.200 0.200 12.111 1.665 0.163 1.451 3.587 3.457 7 72.995 0.430 11.998 1.540 0.251 1.265 3.900 3.698 8 74.023 0.235 12.457 1.568 0.099 1.315 3.547 2.998 Mean 73.434 0.229 , 12.298 1.532 0.143 1.377 3.565 3.396

Sample 7.36 T20/469100 75.829 0.202 13.487 1.672 0.247 1.659 3.692 2.982 2 74.655 0.289 13.383 1.866 0.274 1.678 4.101 2.782 3 74.569 0.216 12.998 1.532 0.257 1.699 3.889 3.204 4 74.587 0.247 13.258 1.665 0.243 1.637 3.952 2.936 5 74.897 0.302 13.498 1.567 0.236 1.521 3.265 2.568 6 75.112 0.256 13.003 1.587 0.213 1.563 3.777 2.886

A8 Appendix 2: Rhyolitic glass

Mean 74.942 0.252 13.271 1.648 0.245 1.626 3.779 2.893

Sample 9.37 T20/468102 72.100 0.242 12.624 1.561 0. 107 1.370 3.353 3.422 2 71 .256 0. 109 12.188 1.317 0. 170 1.193 2.829 3.698 3 71.590 0.172 12.293 1.737 0. 163 1.192 3.086 3.609 4 71 .845 0. 117 12.073 1.605 0.128 1.318 3.389 3.643 5 73.379 0. 192 12.139 2.019 0. 137 1.269 2.775 3.500 6 72.169 0.395 12.118 1.520 0. 167 1.331 3.419 3.452 7 73.057 0. 150 12.329 1.954 0.107 1.214 3.041 3.410 8 72.475 0.000 12.120 2.032 0. 126 1.117 3.290 3.758 9 71.924 0.000 12.019 2.082 0. 112 1.247 3.717 3.680 10 71.750 0.000 12.321 1.549 0. 156 1.431 3.628 3.535 Mean 72. 155 0. 138 12.222 1.738 0. 137 1.268 3.253 3.571

Sample 9.36 T20/468 102 1 72.618 0.208 12.510 1.625 0. 147 1.292 3.126 3.639 2 72.802 0.217 12.418 1.965 0.102 1.253 3.319 3.721 3 73.310 0.190 12.376 1.675 0.132 1.232 3.318 3.642 4 72.724 0.163 12.414 1.226 0.210 1.158 3.011 3.600 5 73.265 0.278 12.567 1.368 0.218 1.386 3.215 3.555 6 74.777 0. 189 13.035 1.333 0.158 1.255 3.215 3.368 Mean 73.249 0.208 12.553 1.532 0.161 1.259 3.201 3.588

Sample 9.35 T20/468102 73.673 0. 192 13.234 1.822 0.268 1.521 3.886 2.636 2 73.254 0. 162 12.280 1.824 0.082 1.307 3.444 3.633 3 74.216 0. 187 13.005 1.668 0.125 1.406 3.446 2.949 4 74.365 0.236 13.458 1.664 0.147 1.456 3.584 3.640 5 73.998 0.205 13.612 1.624 0. 179 1.258 3.845 3.514 6 73.257 0.166 12.998 1.512 0. 166 1.367 2.994 3.446 7 74 .1 11 0. 114 12.359 1.699 0. 135 1.357 3.489 3.495 8 74.369 0. 178 13.547 1.521 0.232 1.266 3.487 3.651 9 75.002 0.154 13.200 1.478 0. 148 1.312 3.265 2.846 10 73. 334 0. 125 12.690 1.588 0.197 1.389 3.658 3.235 Mean 73.958 0. 172 13.038 1.640 0.168 1.364 3.510 3.305

Sample 9.33 T20/468102 1 74.508 0. 164 11.854 1.21 3 0.078 1.095 3.152 3.043 2 72.846 0. 183 11.597 1.064 0.100 1.001 3.189 3.106 3 72.836 0. 173 11.451 1.283 0.201 0.824 2.273 3.430 4 72.440 0. 183 11.280 1.205 0.157 0.688 2.792 3.626 5 73.491 0.109 12.130 1.165 0.131 1.011 3.076 2.785 6 73.162 0. 152 12.085 1.206 0.129 1.050 3.287 2.762 7 73.190 0.097 12.061 1.248 0.157 0.915 3.051 2.942 8 74.178 0.101 12.093 1.335 0.156 1.123 3.038 2.959 9 74.396 0.060 11.659 1.124 0.141 0.970 2.608 2.747 Mean 73.450 0. 136 11.801 1.205 0.139 0.964 2.941 3.044

Sample 9.25 T20/468102 70.521 0.379 12.285 2.180 0.350 1.660 3.019 3.820 2 73.076 0.393 12.531 2.608 0.412 1.446 3.552 3.681 3 73. 144 0.449 12.401 2.663 0.415 1.656 2.886 3.283 4 76.662 0.309 13.279 2.002 0.285 1.393 4. 148 2.796 5 75.783 0.372 13.299 2. 125 0.261 1.662 2.298 2.855 6 77.303 0.275 13.089 1.892 0.223 1.212 4.084 3.026 7 76. 158 0.248 13.041 1.498 0. 176 1.275 4.142 2.745 8 75.974 0.253 13.125 1.637 0. 192 1.295 4.232 2.808 Mean 74.828 0.335 12.881 2.076 0.289 1.450 3.545 3. 127

Sample 9.22 T20/468102 75.663 0.293 13.352 2. 121 0.265 1.554 4.078 2.846 2 72.623 0.241 12.922 2.073 0.299 1.474 3.515 2.638 3 75.509 0.093 12.773 1.965 0.097 1.368 3.353 3.731 4 74.301 0.078 12.382 1.738 0.036 1.174 3.524 3.365 5 74.605 0.405 13.438 2.140 0.335 1.461 4.057 3.013 6 74.716 0.285 13.377 1.992 0.303 1.388 4. 150 2.891 Mean 74.570 0.233 13.041 2.005 0.223 1.403 3.780 3.081

Sample 9.21 T20/468102 1 76.538 0.210 12.141 1.090 0.1 18 1.042 3.361 3.040 2 76.252 0. 167 11.865 1.038 0.134 0.969 3.696 3.044 3 78.179 0. 162 12.245 1.124 0.170 0.875 3.656 3.124 4 75.750 0.137 11.849 1.070 0.192 1.066 2.936 2.710 5 76.087 0.062 11.854 1.251 0.137 0.851 3.333 3.015 6 76.019 0.105 11.427 0.922 0.003 0.607 3.777 3.707 7 75.026 0.169 11.337 0.964 0.043 0.563 3.665 3.718 8 75.729 0.151 11.268 0.975 0.044 0.510 3.647 3.972 Mean 76.198 0.145 11.748 1.054 0.105 0.810 3.509 3.291

Sample 14.23 T20/4 771 12 76.213 0.328 13.295 2.025 0.298 1.504 3.699 2.830

A9 Appendix 2: Rhyolitic glass

2 75.990 0.320 13.343 1.918 0.290 1.971 3.808 2.709 3 75.515 0.345 13.110 1.930 0.264 1.494 3.914 3.239 4 74. 176 0.134 12.690 1.643 0.113 1.227 3.206 2.785 5 74.758 0.246 12.528 1.499 0.128 1.209 3.418 2.710 6 74.677 0.145 12.899 1.589 0.163 1.338 3.712 2.933 7 76.197 0.313 13.750 1.942 0.293 1.663 3.951 2.800 8 76.205 0.228 13.093 2.266 0.258 1.584 3.929 2.801 Mean 75.466 0.257 13.089 1.852 0.226 1.499 3.705 2.851

Sample 14.22 T20/477112 72.115 0.124 11.262 1.086 0.162 1.081 3.021 2.824 2 74.828 0.152 11.671 1.197 0.144 1.009 3.004 2.975 3 75.123 0.145 11.833 0.949 0.150 1.078 3.222 3.001 4 74.408 0. 102 11.849 0.885 0.070 0.960 3.290 3.016 5 76.287 0. 137 12.317 1.082 0. 116 1.038 2.542 3.047 6 74.345 0.165 11.968 1.114 0.118 0.887 2.676 2.837 7 77.982 0.121 12.506 1.217 0.184 1.048 3.582 3.196 Mean 75.013 0.135 11.915 1.076 0.135 1.014 3.048 2.985

A 10 Appendix 3.1 : Mineralogy

APPENDIX 3: ANDESITIC TEPHRAS

3.1 Ferromagnesian mineral assemblages

The following table presents modal ferromagnesian mineralogy of all the andesitic tephras sampled in this study. Counts of 400 grains were made on mineral separates.

Sample Location Section Unit or correlation OPX CPX TM HB OL

93.23 T20/461 096 45 56 39 5 0 0 93.24 T20/464098 46 59 40 1 0 0 93.25 41 37 3 0 19 93.26 Marker Unit 1 8 36 1 0 56 93.27 Marker Unit 1 16 48 1 0 36 93.28 T20/465099 47 51 48 0 0 2 93.29 T20/469102 9 57 43 0 0 0 93.30 T20/466096 5 54 45 0 0 0 93.31 47 51 0 0 2 93.34 T20/469100 7 50 49 0 0 1 93.36 53 47 0 0 0 93.38 59 34 0 6 93.39 Marker Unit 2 39 50 10 2 0 93.40 56 44 0 0 0 93.41 T20/468102 9 44 54 0 0 2 93.42 47 52 1 0 0 93.43 57 34 9 0 0 93.44 58 33 9 0 0 93.45 57 29 14 0 0 93.46 51 31 18 0 0 93.47 Marker Unit 3 49 40 11 0 0 93.48 63 22 15 0 0 93.49 67 19 15 0 0 93.50 66 20 15 0 0 93.51 52 43 4 1 0 93.52 T20/468102 10 Marker Unit 4 54 39 6 1 0 93.53 59 41 1 0 0 93.54 55 40 2 0 3 93.55 55 42 3 1 0 93.56 58 41 1 0 0 93.57 59 29 12 0 0 93.58 72 15 13 0 0 93.60 T20/4771 12 14 49 51 1 0 0 93.61 (93.37) 49 43 9 0 0 93.62 (93.38) 65 28 0 6 93.63 (93.40) 58 41 0 1 93.64 (93.40) 47 52 0 93.65 (93.42) 54 44 1 0 2 93.66 65 29 5 0 1 93.67 53 43 4 0 1 93.68 68 22 10 0 1 93.69 66 26 8 0 0 93.70 Marker Unit 3 57 41 2 0 0 93.71 (93.48) 66 28 6 0 0 93.72 (93.49) 57 27 16 0 0 93.73 (93.51) 49 40 8 2 1 93.74 Marker Unit 4 56 39 5 0 0 93.75 52 47 1 0 0 93.76 56 42 2 0 0 93.77 62 27 11 0 0 93.78 75 12 14 0 0 93.79 59 40 2 0 0

A 11 Appendix 3.1 : Mineralogy

93.80 61 30 9 0 0 93.81 63 34 3 0 0 93.82 43 56 2 0 1 93.83 49 50 1 0 0 93.84 73 22 5 0 0 93.85 Marker Unit 5 50 41 9 0 1 93.86 51 37 12 0 0 93.87 61 36 4 0 0 93.88 60 37 3 0 0 93.89 Marker Unit 6 27 33 9 31 0 93.91 Marker Unit 7 55 39 1 5 0 94.2 T20/455088 Bullot Fm. 68 24 8 0 0 94.4 T20/469102 9 1 Bullot Fm. 69 28 3 0 0 94.13 T20/496157 51 L 76 20 4 0 0 94.15 T20/463097 Bullot Fm. 47 10 15 28 0 94.16 78 13 9 0 0 94.17 66 26 8 0 0 94.18 T20/498184 53 ?Bullot Fm. 66 24 9 1 0 94.20 T19/484209 54 71 22 4 0 3 94.22 T19/481211 55 {94.31) 36 31 7 26 0 94.31 T19/495220 59 {94.22) 37 14 9 40 0 94.32 T20/477112 14 Marker Unit 7 20 6 8 66 0 94.41 T19/491239 Rotoaira Tephra 78 19 3 0 0 94.43 T19/496238 66 Bullot Fm. 71 28 <1 <1 0 94.44 75 25 <1 0 0 94.45 72 24 4 <1 0 94.46 {94.22&31) 39 34 27 0 0 94.79 T19/364544 73 Poutu Tephra 63 29 7 0 1

A 12 Appendix 3.2: Hornblende analyses

3.2 Hornblende analyses

All analyses are listed in a raw form (not normalised).

Sample Location Section Analysis Si02 Ti02 Al203 FeO M nO MgO CaO Na20 K20

Sample 93.39 T20/469100 7 44.665 1.192 13.032 13.215 0.171 15.805 10.682 2.840 0.217 2 38.562 2.272 12.035 12.349 0.421 12.403 11.561 2.078 0.768 3 39.565 2.225 12.821 12.959 0.348 12.773 11.694 2.161 0.844 Mean 40.931 1.896 12.629 12.841 0.313 13.660 11.312 2.360 0.610

Sample 93.48 T20/468102 9 1 39.885 3.106 11.854 12.020 0.402 13.520 11.169 2.428 0.951 2 38.493 2.965 11.211 12.913 0.361 12.903 11.270 2.325 0.890

Sample 93.51 T20/468102 9 41 .877 2.039 13.772 11.474 0.089 14.617 10.957 2.398 0.415 2 42.615 1.941 13.611 12.109 0.152 14.819 10.824 2.438 0.360 3 41.589 1.759 13.368 12.598 0.255 13.932 10.691 2.358 0.382 4 41 .504 1.710 13.048 11.264 0.318 14.330 10.723 2.410 0.430 5 42.194 2.126 13.976 11.608 0.420 14.762 11.128 2.433 0.331 6 40.067 2.054 13.269 11.321 0.260 13.863 10.574 2.238 0.411 7 41 .245 3.305 11.261 11.856 0.333 14.673 11.192 2.546 0.815 8 41 .473 3.301 11.707 11.459 0.206 14.344 11.089 2.524 0.915 9 41 .103 3.546 11.620 11.116 0.524 14.282 11.272 2.269 0.893 Mean 41.519 2.420 12.848 11.645 0.284 14.402 10.939 2.402 0.550

Sample 93.52 T20/468102 10 1 42.820 1.168 12.634 13.569 0.180 14.161 11.084 2.040 0.225 2 42.945 1.228 12.232 14.141 0.387 13.373 10.775 1.976 0.295 3 43.290 1.261 11.770 14.338 0.579 13.823 10.424 1.913 0.263 4 42.051 1.215 12.006 14.515 0.172 13.527 10.602 2.014 0.173 5 41.725 1.194 12.094 13.928 0.369 13.407 10.500 1.920 0.341 6 40.830 1.199 12.079 13.755 0.318 13.093 10.592 1.873 0.325 Mean 42.277 1.21 1 12.136 14.041 0.334 13.564 10.663 1.956 0.270

Sample 93.55 T20/468102 10 35.407 2.973 11.795 12.297 0.258 11.878 12.190 2.258 0.936 2 37.31 1 3.369 11.968 12.102 0.217 11.708 12.321 2.420 0.975 3 39.803 2.930 13.340 12.406 0.229 13.137 11.985 2.542 1.039 Mean 37.507 3.091 12.368 12.268 0.235 12.241 12.165 2.407 0.983

Sample 93.71 T20/477112 14 39.540 2.967 9.925 11.727 0.482 13.630 10.293 2.187 0.758

Sample 93.73 T20/477112 14 1 40.694 2.110 13.755 11.241 0.187 14.492 11.052 2.300 0.387 2 37.426 2.153 12.407 11.705 0.313 12.793 10.885 1.951 0.297 3 40.770 2.038 13.368 12.032 0.297 14.674 10.835 2.349 0.348 4 41 .085 2.018 13.117 11.816 0.282 14.543 10.556 2.327 0.340 5 39.621 1.883 13.057 11.675 0.192 14.477 10.633 2.095 0.371 6 40.477 2.016 13.542 10.875 0.196 14.468 11.063 2.300 0.398 7 41 .163 1.990 13.481 11.900 0.342 14.677 11.266 2.292 0.350 8 40.863 2.062 13.577 11.505 0.451 14.756 11.013 2.445 0.308 9 40.818 2.034 13.220 11.417 0.306 14.555 10.961 2.357 0.317 10 40.594 2.169 13.580 12.018 0.420 14.637 10.795 2.408 0.313 11 40.453 3.109 12.517 12.955 0.355 13.446 10.989 2.561 0.844 12 40.595 2.010 13.709 11.938 0.266 14.666 10.975 2.669 0.324 Mean 40.380 2.133 13.278 11.756 0.301 14.349 10.919 2.338 0.383

Sample 93.75 T20/477112 14 1 40.764 2.269 11.681 11.717 0.467 13.886 11.787 2.425 0.749 2 42.518 2.149 11.900 12.145 0.372 14.661 11.948 2.690 0.743 3 41 .933 3.085 12.150 11.366 0.297 14.269 11.426 2.638 0.730 4 43.624 3.262 12.457 11.954 0.312 14.683 11.341 2.659 0.759 Mean 42.210 2.691 12.047 11.796 0.362 14.375 11.626 2.603 0.745

Sample 93.89 T20/477112 14 1 41.637 2.180 13.451 11.151 0.188 14.937 12.239 2.565 0.856 2 41 .062 2.790 12.499 12.785 0.279 13.272 11.854 2.493 0.884 3 40.784 2.014 13.681 11.425 0.121 13.779 12.467 2.358 0.933 4 36.875 2.132 13.902 12.434 0.151 11.961 12.361 2.332 0.810 5 41 .979 2.787 11.532 13.134 0.373 13.283 12.201 2.625 0.827 6 41 .056 2.109 13.644 14.557 0.380 12.778 12.392 2.607 0.764 7 40.159 1.971 14.230 13.105 0.206 12.848 12.541 2.487 0.928 8 41 .546 2.490 12.084 13.113 0.423 12.681 11.795 2.665 0.766 9 44.438 3.081 10.475 11.446 0.421 14.632 11.983 2.395 0.759 10 42.613 1.907 13.347 10.551 0.212 14.955 12.752 2.580 0.957 Mean 41.215 2.346 12.885 12.370 0.275 13.513 12.259 2.51 1 0.648

Sample 93.90 T20/4771 12 14 41.845 1.743 13.223 12.900 0.250 13.661 11.295 2.644 2 38.954 1.850 11.738 13.529 0.242 12.319 11.594 1.855 0.70.66799

A 13 Appendix 3.2: Hornblende analyses

3 39.785 2.719 8.082 12.095 0.501 13.193 11.323 1.659 0.790 4 38.381 2.579 12.453 13.656 0.326 11.983 11.090 2.615 0.733 Mean 39.741 2.223 11.374 13.045 0.330 12.789 11.326 2. 193 0.747

Sample 93.91 T20/477112 14 1 41 .030 2.489 11.101 12.323 0.375 13.477 11.819 2.454 0.648 2 43.698 2. 918 10.792 11.713 0.504 14.527 11.910 2.570 0.739 3 42.780 3. 122 11.802 12.194 0.507 14.053 11.713 2.623 0.748 4 41 .701 2.224 14.352 11.780 0.173 14.632 12.104 2.545 0.938 5 42.739 2.306 14.220 12.205 0.145 14.032 12.733 2.508 0.890 6 40.861 2.281 14.028 12.880 0.323 13.433 12.594 2.444 0.832 7 42.703 3.070 12.369 13.065 0.519 13.625 12.435 2.872 0.907 8 40.427 2.055 12.999 11.021 0.265 14.956 12.573 2.563 0.921 9 43.738 2.012 14.289 10.122 0.232 16.407 11.372 2.885 0.886 10 39.626 2.050 13.034 10.459 0.063 14.600 12.262 2.377 0.904 Mean 41 .930 2.453 12.899 11.776 0.311 14.374 12.152 2.584 0.841

Sample 94. 15 T20/463097 52 1 43.937 2.431 10.585 11.745 0.202 14.499 10.975 2.230 0.277 2 45.061 2.458 11.108 12.158 0.245 14.998 10.960 2.279 0.333 3 45.070 2.143 11.218 12.341 0.224 15.053 10.931 2.289 0.335 4 43.937 2.314 10.900 12.258 0.238 14.524 11.183 2.044 0.388 5 44. 144 2.432 11.809 12.815 0. 163 14.491 10.953 2.349 0.361 Mean 44.430 2. 356 11.124 12.263 0.214 14.713 11.000 2.238 0.339

Sample 94. 18 T20/498184 53 1 41 .015 2. 162 13.023 12.125 0.097 14.446 10.924 2.208 0.329 2 41 .241 2. 170 13.087 12.021 0.21 1 14.655 11.305 2.209 0.408 3 40.385 1.790 12.656 10.906 0.261 14.471 10.957 2.197 0.356 4 41 .797 2.074 13.057 12.513 0. 199 14.526 10.978 2.391 0.331 Mean 41 .110 2.049 12.956 11.891 0. 192 14.525 11.041 2.251 0.356

Sample 94.22 T1 9/48121 1 55 46.023 1.746 9.129 13.958 0.407 13.497 10.918 1.765 0.322 2 44.588 1.685 9.607 14.838 0.293 13.671 10.876 1.812 0.296 3 45.905 1.341 8.542 13.606 0.355 14.116 10.331 1.400 0.416 4 44.626 1.422 9.156 13.824 0.369 13.607 10.190 1.643 0.337 5 45.176 1.801 9.575 13.166 0.268 14.016 10.948 1.646 0.363 6 47.106 1.299 7.443 12.733 0.280 15.729 10.568 1.281 0.244 Mean 45.571 1.549 8.909 13.688 0.329 14.106 10.639 1.591 0.330

Sample 94.31 T1 9/495220 59 44.845 1.204 11.357 13.874 0.484 13.982 10.677 1.948 0.219 2 43.904 2.289 10.810 13.469 0.276 13.91 1 10.719 2.243 0.321 3 44.054 1.319 12.312 13.677 0.306 13.689 10.748 2.420 0.380 4 44.429 1.613 10.548 11.995 0.312 14.086 10.938 2.191 0. 333 5 42.430 1.402 11.312 12.043 0.207 14.128 10.827 2.093 0.276 6 44. 191 1.421 12.248 11.005 0.214 14.849 10.955 2.258 0.340 Mean 43.976 1.541 11.431 12.677 0.300 14.108 10.81 1 2. 192 0.312

Sample 94.43 T1 9/496238 66 43.372 1.195 12.530 14.91 7 0.948 13.144 10.51 1 1.932 0.381 2 41.124 1.306 12.446 15.285 0.509 12.246 10.356 1.793 0.370 3 40.582 1.523 12.809 12.933 0.354 13.105 11.181 2. 122 0.310 4 40.934 1.686 13.726 13.974 0.183 13.499 11.082 2.401 0.337 5 41 .973 1.664 9.901 14.280 0.377 13.363 10.223 1.971 0.233 6 41 .486 1.416 10.680 13.746 0.455 13.325 10.705 1.907 0.336 Mean 41 .579 1.465 12.015 14.189 0.471 13.114 10.676 2.021 0.328

Sample 94.45 T19/496238 66 1 41 .092 1.428 13.813 12.641 0.329 14.281 10.903 2.416 0.324 2 42.314 1.553 13.244 13.143 0. 182 14.479 11.171 2.173 0.278 3 43. 101 1.594 13.083 12.169 0.260 14.935 11.114 2.264 0.203 4 41.419 1.645 13.638 11.396 0.141 14.876 10.815 2.31 1 0.293 5 44.565 0.958 9.896 16.091 0.681 13.331 10.261 1.585 0.375 6 42.369 1.256 12.675 11.644 0.135 15.502 10.793 2.141 0.244 7 41 .706 1.464 14.105 12.071 0.274 14.226 10.804 2.422 0.285 Mean 42.367 1.414 12.922 12.736 0.286 14.519 10.837 2. 187 0.286

Sample 94.46 T19/496238 66 1 40.721 2.851 11.133 10.830 0.252 14.487 11.597 2.368 0.806 2 41 .669 2.610 11.317 11.679 0.316 15.010 11.479 2.534 0.633 3 41 .659 2.81 1 11.113 11.513 0.254 14.513 11.326 2.293 0.837 4 41 .925 2.974 11.051 10.917 0.233 14.910 11.282 2.426 0.795 5 41.017 2.528 11.049 11.109 0. 334 14.401 11.340 2.342 0.701 6 40.452 2.476 11.486 12.258 0.324 13.877 11.210 2.1 18 0.738 7 43.502 3.011 11.659 11.712 0.450 14.677 11.333 2.521 0.774 Mean 41 .564 2.752 11.258 11.431 0.309 14.554 11.367 2.372 0.755

Sample 94.47 T19/496238 66 1 45.625 1.330 6.683 15.81 1 0.531 13.287 10.166 1.531 0.303 (Rhyolitic) 2 45.112 1.393 6.369 16.534 0.604 12.483 10.132 1.436 0.327 3 44.111 1.865 7.853 14.480 0.474 13.609 10.569 1.539 0.220 Mean 44.949 1.529 6.968 15.608 0.536 13.126 10.289 1.502 0.283

A14 Appendix 3.3: Olivine analyses

3.3 Olivine analyses

All analyses are listed in a raw form (not normalised).

Sample Location Section Analysis Si02 Ti02 Al203 FeO M nO M gO CaO Na20 1<20

Sample 93.25 T20/464098 46 1 resorbed 37.947 0.012 0.000 21 .649 0.190 39.953 0.1 19 0.005 0.029 2resorbed 39.090 0.000 0.007 18.170 0.125 43.692 0.133 0.058 0.026

39.674 0.062 0.008 11.806 0.284 47.596 0.186 0.050 0.035 2 40.235 0.100 0.035 12.237 0.331 47.465 0.100 0.035 0.109 3 40.832 0.029 0.049 15.408 0.260 46.610 0.153 0.051 0.024 4 41 .481 0.029 0.015 12.442 0.266 48.663 0.177 0.049 0.051 5 41 .462 0.021 0.052 12.876 0.228 48. 540 0.133 0.005 0.020 Mean 40.737 0.048 0.032 12.954 0.274 47.775 0.150 0.038 0.048

Sample 93.26 T20/464098 46 39.200 0.021 0.097 14.765 0.529 45.887 0.152 0.031 0.002 2 39.51 2 0.034 0.050 14.514 0.468 45.969 0. 178 0.090 0.010 3 39.394 0.070 0.083 15.626 0.288 45.407 0. 118 0.013 0.025 4 38.805 0.053 0.014 15.306 0.159 45.285 0. 146 0.057 0.039 5 39.633 0.053 0.045 14.676 0.027 46 .170 0. 180 0.035 0.044 6 39.686 0.009 0.055 13.826 0.085 45.844 0.149 0.029 0.053 7 39.212 0.245 0.018 14.872 0.105 45.598 0. 167 0.002 0.087 8 39.354 0.066 0.01 5 14.816 0.326 44.984 0.159 0.068 0.035 9 39.689 0.074 0.061 14.402 0.071 46.199 0. 129 0.001 0.043 10 39.370 0.039 0.019 15.136 0.019 45.900 0.128 0.005 0.029 Mean 39.386 0.066 0.046 14.794 0.208 45.724 0.151 0.033 0.037

Sample 93.27 T20/464098 46 39. 120 0.040 0.015 14.819 0.509 45.762 0.158 0.002 0.01 5 2 39.323 0.057 0.059 15.647 0.393 45.808 0.160 0.000 0.031 3 38.974 0.027 0.066 14.613 0.319 45.413 0.218 0.044 0.037 4 39.985 0.000 0.074 13.881 0.466 46.403 0.205 0.031 0.008 5 39.735 0.001 0.079 14.988 0.301 45.332 0.195 0.005 0.01 7 6 40.290 0.000 0.051 14.349 0.659 46.067 0.141 0.037 0.026 7 40.088 0.000 0.115 14.481 0.384 46.105 0.188 0.001 0.031 8 39.963 0.000 0.057 15.493 0.528 40.089 0.145 0.008 0.076 9 39.439 0.094 0.059 14.013 0.252 46.607 0.155 0.000 0.043 10 39.308 0.005 0.039 15.836 0.370 45.153 0.166 0.000 0.001 Mean 39.623 0.022 0.061 14.812 0.418 45.274 0.173 0.013 0.029

Sample 93.28 T20/465099 47 1 resorbed 37.820 0.049 0.000 23.829 0.430 38.142 0.146 0.005 0.051 2resorbed 35.922 0. 000 0.019 23.503 0.391 37.540 0.158 0.000 0.059

1 38.018 0.000 0.013 18.166 0.272 43.807 0.130 0.083 0.027 2 38.001 0.056 0.1 16 18.618 0.174 42.628 0.139 0.013 0.032

Sample 93.31 T20/466096 5 38.200 0.012 0.063 21.276 0.358 39.613 0.156 0.037 0.003 2 38.157 0.002 0.076 20.525 0.318 40.809 0.142 0.037 0.017 3 38.036 0.000 0.043 19.359 0.378 40.925 0.100 0.071 0.000 4 35.920 0.070 0.077 20.715 0.304 37.892 0.156 0.239 0.095 Mean 37.578 0.021 0.065 20.469 0.340 39.810 0.139 0.096 0.029

Sample 93.34 T20/469 100 7 1 39.006 0.078 0.051 16.481 0.313 47.662 0. 104 0.000 0.009 2 39.444 0.031 0.015 15.039 0.284 46.085 0.124 0.000 0.041 3 39.159 0.030 0.018 15.085 0.264 47.255 0.149 0.000 0.012 4 40.288 0.058 0.036 12.968 0.209 46.560 0.157 0.011 0.006 Mean 39.474 0.049 0.030 14.893 0.268 46.891 0.134 0.003 0.017

Sample 93.38 T20/469100 7 1 resorbed 37.827 0.055 0.106 25.870 0.622 36.703 0.159 0. 113 0.053

41 .053 0. 123 0.042 9.753 0.256 49.300 0.152 0.013 0.039 2 40.307 0.061 0.060 11.468 0.260 47.998 0.199 0.029 0.053 3 39.419 0.005 0.067 13.450 0.263 46.319 0.154 0.000 0.026 4 40.048 0.048 0.012 11.061 0.490 48.455 0.191 0.000 0.027 5 39.967 0.013 0.016 16.007 0.343 45. 334 0.109 0.089 0.01 1 Mean 40 .159 0.050 0.039 12.348 0.322 47.481 0.161 0.026 0.031

Sample 93.41 T20/468102 9 39.264 0.000 0.018 12.556 0.200 45.544 0.137 0.028 0.024 2 39.234 0.020 0.088 16.1 18 0.265 43. 130 0.137 0.000 0.037 3 39. 152 0.048 0.01 7 15.840 0.303 44.091 0.132 0.000 0.000 4 38.377 0.075 0.132 18.183 0.262 41.520 0.100 0.007 0.025 5 39.733 0.000 0.01 7 16.611 0. 182 44.058 0.181 0.086 0.048 Mean 39. 152 0.029 0.054 15.862 0.242 43.669 0.137 0.024 0.027

A 15 Appendix 3.3: Olivine analyses

Sample 93.44 T20/468102 9 1 resorbed 38.366 0.016 0.035 26.864 0.429 35.010 0.087 0.003 0.031 2resorbed 38.510 0.047 0.000 26.965 0.532 35.077 0.069 0.001 0.047

1 40.593 0.032 0.056 17.205 0.346 43.407 0.139 0.005 0.016 2 40.028 0.039 0.01 1 17.300 0.295 43.016 0.098 0.056 0.041

Sample 93.54 T20/468 102 10 38.973 0.000 0.016 13.097 0.231 46.030 0.167 0.01 1 0.023 2 39.854 0.000 0.208 12.091 0.203 47.247 0.141 0.085 0.010 3 39.154 0.01 1 0.014 13.924 0.199 45.506 0.144 0.058 0.009 4 38.293 0.086 0.095 17.940 0.268 41.761 0.122 0.078 0.007 Mean 39.069 0.024 0.083 14.263 0.225 45. 136 0.144 0.058 0.012

Sample 93.60 T20/477112 14 1 39.254 0.030 0.009 12.403 0.224 47.222 0.166 0.070 0.037 2 39.982 0.075 0.012 12.225 0.277 47.632 0.135 0.073 0.043 3 39.828 0.128 0.062 12.621 0.197 47.547 0.196 0.139 0.045

4 38.610 0.061 0.088 18.547 0.252 43.273 0.180 0.104 0.064 5 38.971 0.048 0.047 16.325 0.269 43. 140 0.146 0.070 0.043

Sample 93.62 T20/4771 12 14 1 38.619 0.064 0.015 8.500 0.136 51.163 0.156 0.067 0.052 2 38.232 0.000 0.069 8.496 0.117 52.488 0.131 0.012 0.047

3 36.979 0.051 0.067 13.970 0.171 46.301 0.114 0.056 0.002 4 37.459 0.039 0.055 14.115 0.298 47.498 0.169 0.092 0.000 5 37.612 0.025 0.010 13.846 0.260 47.395 0.147 0.055 0.018 6 36.863 0.115 0.001 11.823 0.313 49.300 0.181 0.058 0.027 Mean 37.228 0.058 0.033 13.439 0.261 47.624 0.153 0.065 0.012

Sample 93.63 T20/4771 12 14 36.203 0.123 0.045 22.252 0.298 39.614 0.099 0.065 0.044

2 38.549 0.005 0.046 14.057 0.116 47.669 0.112 0.000 0.004 3 36.820 0.002 0.000 13.757 0.263 46.787 0.124 0.000 0.002 4 39.259 0.028 0.000 11.022 0.190 49.080 0.125 0.01 1 0.027

Sample 93.65 T20/4771 12 14 39.778 0.091 0.074 14.115 0.183 46.203 0.081 0.066 0.018 2 40.818 0.112 0.069 14.191 0.149 46.535 0.113 0.013 0.000 3 40.050 0.046 0.000 11.890 0.209 47.921 0.135 0.000 0.007 4 39.856 0.033 0.000 11.205 0.167 48.208 0.092 0.000 0.016 Mean 40.126 0.071 0.036 12.850 0.177 47.217 0.105 0.020 0.010

Sample 93.67 T20/477112 14 1 38.507 0.010 0.000 18.549 0.210 43. 155 0.100 0.101 0.026 2 39.563 0.057 0.040 17.745 0.192 42.841 0.124 0.077 0.035 3 36.638 0.068 0.018 18.131 0.329 42.189 0.138 0.000 0.042 4 37.750 0.054 0.059 17.386 0.314 42.863 0.133 0.101 0.030 Mean 38.115 0.047 0.029 17.953 0.261 42.762 0.124 0.070 0.033

Sample 93.68 T20/477112 14 1 40.023 0.015 0.065 17.973 0.371 43.747 0.123 0.050 0.018 2 39.422 0.052 0.168 17.853 0.252 45.803 0.421 0.030 0.000 3 39.460 0.024 0.01 1 17.086 0.353 44.393 0.150 0.084 0.023 4 37.240 0.01 1 0.074 19.531 0.346 43.195 0.136 0.073 0.022 Mean 39.036 0.026 0.080 18.111 0.331 44.285 0.208 0.059 0.016

Sample 93.70 T20/477112 14 1 36.074 0.254 0.293 29.819 0.318 30.813 0.309 0.053 0.031 2 35.435 0.120 0.014 31.139 0.536 28.757 0.283 0.004 0.046 3 35.566 0.069 0.516 29.289 0.636 29.271 0.283 0.402 0.041 4 35.989 0.060 0.000 27.177 0.625 34.342 0.187 0.084 0.054 5 36.485 0.032 0.057 27.146 0.584 34.285 0.393 0.200 0.084 6 34.994 0.122 0.204 31.220 0.466 27.644 0.317 0.168 0.024 Mean 35.757 0.110 0.181 29.298 0.528 30.852 0.295 0.152 0.047

Sample 93.71 T20/477112 35.369 0.100 0.563 31.548 0.341 27.134 0.356 0.084 0.036 2 35.460 0.136 0.430 31 .048 0.51 1 27.458 0.378 0.095 0.030 3 35.681 0.111 0.251 29.746 0.506 30.066 0.312 0.138 0.044 4 35.256 0.1 12 0.325 31.056 0.458 29.687 0.310 0.106 0.021 5 35.487 0.105 0.465 29.998 0.347 29.874 0.398 0.068 0.059 6 35.540 0.121 0.402 30.157 0.503 29.325 0.398 0.098 0.041 Mean 35.466 0.114 0.406 30.592 0.444 28.924 0.359 0.098 0.039

Sample 93.77 T20/477112 14 1 41.438 0.152 0.017 13.766 0.247 46.510 0.183 0.037 0.053 2 41.302 0.025 0.015 13.691 0.265 46.839 0.191 0.045 0.057 3 40.576 0.012 0.069 12.016 1.973 46.598 0.164 0.042 1.528

Sample 93.82 T20/477112 14 36. 179 0.029 0.154 18.471 0.434 42.790 0.238 0.088 0.151

A 16 Appendix 3.3: Olivine analyses

2 34.925 0.028 0.040 19.079 0.335 42.308 0.151 0.071 0.055 3 33.552 0.01 1 0.098 21 .213 0.298 39.985 0.149 0.091 0.017

Sample 93.85 T20/4771 12 14 1 36.119 0.014 0.108 24.502 0.436 37.754 0.066 0.145 0.041 2 35.601 0.035 0.035 24.029 0.428 38.989 0.101 0.047 0.048 3 35.985 0.014 0.098 23.684 0.425 38.647 0.095 0.102 0.045 4 35.998 0.016 0.112 23.268 0.462 38.654 0.035 0.078 0.044 5 36.002 0.025 0.105 24. 125 0.465 38.264 0.037 0.089 0.046 6 35.654 0.056 0.065 24.517 0.425 37.724 0.105 0.067 0.048 7 35.458 0.023 0.099 24.056 0.402 38.025 0.095 0.098 0.041 8 36.015 0.012 1.070 24.265 0.414 37.951 0.102 0.068 0.039 Mean 35.854 0.024 0.212 24.056 0.432 38.251 0.080 0.087 0.044

A 17 Appendix 3.4: Titanomagnetite analyses

Appendix 3.4 Titanomagnetite analyses

All analyses are in a raw form (not normalised).

Sample Location Section Analysis Si02 Ti02 Al203 FeO MnO MgO CaO Na20 K20 Cr203 NiO

Sample 93.24 T20/464098 46 1 0.064 11.691 3.046 72.650 0.285 2.823 0.095 0.239 0.052 0.776 0.138 2 0.150 10.532 3.181 75.910 0.188 1.415 0.148 0.300 0.000 0.081 0.047 3 0.290 12.030 3.351 71.196 0.329 3.233 0.166 0.359 0.014 0.569 0.176 4 0.059 11.042 3.216 72.686 0.191 3.152 0.046 0.092 0.069 0.081 0.131 5 O.o76 11.383 3.161 71 .688 0.277 2.961 0.058 0.051 0.003 0.293 0.199 Mean 0.128 11.336 3.191 72.826 0.254 2. 717 0.103 0.208 0.028 0.360 0.138

Sample 93.25 T20/464098 46 0.066 13.254 2.506 74.933 0.381 3.387 0.073 0.002 0.023 0.229 0.004 2 0.082 11.627 2.232 79.062 0.658 2.204 0.083 0.227 0.029 0.444 0.106 3 0.128 13.372 2.251 75.805 0.347 3.686 0.048 0.175 0.050 0.120 0.000 4 0.097 15.027 2.373 76.633 0.309 2.452 0.043 0.089 0.036 0.215 0.065 5 0.121 15.125 2.730 73.923 0.456 3.502 0.070 0.006 0.025 0.130 0.064 6 0.125 10.717 3.335 77.494 0.377 3.090 0.027 0.012 0.029 0.260 0.037 7 0.229 11.988 2.262 77.921 0.425 2.275 0.141 0.146 0.025 0.160 0.078 8 0.159 14.266 2.368 74.968 0.433 2.778 0.021 0.088 0.014 0.167 0.056 Mean 0.126 13.172 2.507 76.342 0.423 2.922 0.063 0.093 0.029 0.216 0.051

Sample 93.26 T20/464098 46 1 0.158 8.125 2.501 80.265 0.564 1.025 0.088 0.112 0.045 0.215 0.062 2 0.256 8.698 2.013 80.999 0.441 1.115 0.078 0.165 0.088 0.258 0.098 3 0.142 9.003 1.854 80.698 0.235 1.698 0.089 0.188 0.068 0.232 0.254 4 0.1 11 8.995 1.764 81.254 0.336 1.995 0.056 0.165 0.035 0.247 0.158 5 0.198 8.269 1.846 80.005 0.289 1.654 0.021 0.121 0.056 0.332 0.114 6 0.225 8.784 1.655 80.568 0.364 1.567 0.091 0.133 0.130 0.395 0.102 Mean 0.182 8.646 1.939 80.632 0.372 1.509 0.071 0.147 0.070 0.280 0.131

Sample 93.27 T20/464098 46 1 0.144 9.052 2.001 81.051 0.429 1.581 0.091 0.083 0.019 0.065 0.066 2 0.388 8.458 1.451 80.574 0.407 1.803 0.067 0.138 0.041 0.722 0.000 3 0. 717 8.357 0.876 81.255 0.324 1.225 0.094 0.134 0.035 0.205 0.174 4 0.298 8.671 1.687 81 .600 0.322 1.665 0.099 0.143 0.440 0.360 0.191 5 0.265 8. 798 2.087 80.998 0.345 1.582 0.068 0.152 0.029 0.265 0.115 6 0.154 9.045 2.258 80.254 0.399 1.396 0.024 0.199 0.023 0.030 0.098 Mean 0.328 8.730 1.727 80.955 0.371 1.542 0.074 0.142 0.098 0.275 0.107

Sample 93.38 T20/469100 7 0.065 8.752 4.222 75.048 0.385 3.946 0.087 0.065 0.000 0.920 0.170 2 0.042 8.928 4.504 74.376 0.274 4.050 0.069 0.000 0.021 1.322 0.105 3 0.107 9.646 3.834 77.449 0.346 3.083 0.042 0.000 0.031 0.632 0.055 4 0.086 9.545 4.277 77.194 0.429 3.191 0.046 0.209 0.033 0.521 0.189 5 0.144 9.569 4.224 77.703 0.418 2.853 0.046 0.127 0.010 0.476 0.057 Mean 0.089 9.288 4.212 76.354 0.370 3.425 0.058 0.080 0.019 0.774 0.115

Sample 93.39 T20/469100 7 1 0.102 15.260 1.201 73.025 0.531 2.160 0.052 0.000 0.018 0.153 0.051 2 0.050 15.897 1.115 73.254 0.315 1.882 0.028 0.080 0.050 0.175 0.093 3 0.020 15.555 0.706 73.598 0.225 1.890 0.060 0.057 0.000 0.068 0.000 4 0.136 15.334 1.327 73.300 0.242 1.160 0.042 0.000 0.042 0.141 0.023 5 0.216 15.722 1.241 74.218 0.474 1.258 0.038 0.000 0.023 0.089 0.033 6 0.059 16.002 1.315 72.598 0.256 1.357 0.025 0.057 0.015 0.155 0.048 Mean 0.097 15.628 1.151 73.332 0.341 1.618 0.041 0.032 0.025 0.130 0.041

Sample 93.43 T20/468102 9 1 0.096 13.935 2.652 71 .589 0.510 2.720 0.106 0.188 0.000 0.178 0.034 2 0.096 14.193 2.773 71.559 0.419 2.697 0.041 0.008 0.000 0.394 0.091 3 0.157 14.225 2.765 71.904 0.415 2.643 0.067 0.119 0.000 0.349 0.094 4 0.127 14.684 2.602 72.474 0.458 2.381 0.042 0.097 0.000 0.476 0.095 Mean 0.1 19 14.259 2.698 71.882 0.451 2.610 0.064 0.103 0.000 0.349 0.079

Sample 93.44 T20/468 102 9 1 0.165 9.688 3. 733 74.084 2.783 3.382 0.055 0.008 0.770 0.085 0.165 2 0.21 1 10.425 3.843 73.270 0.403 3.214 0.008 0.162 0.016 0.260 0.198 3 0.233 10.363 3.946 73.948 0.334 3.235 0.032 0.175 0.154 0.202 0.139 4 0.162 10.049 4. 172 75.763 0.388 3.157 0.051 0.073 0.048 0.123 0.155 5 0.261 10.005 3.855 72.278 0.258 3. 338 0.084 0.149 0.037 0.041 0.147 6 0.168 9.801 3.979 75.587 0.323 3.260 0.056 0.191 0.013 0.131 0.162 Mean 0.200 10.055 3.921 74.155 0.748 3.264 0.048 0.126 0.173 0.140 0.161

Sample 93.45 T20/468102 9 0.129 7.472 6.856 76.346 0.222 4.255 0.043 0.158 0.023 0.301 0.006 2 0.132 7.468 7.068 77.093 0.312 3.815 0.037 0.000 0.047 0.850 0.355

3 0.194 12.341 3.271 78.028 0.324 2.868 0.065 0.254 0.009 0.122 0.098 4 0.153 12.037 3.228 75.130 0.398 3.006 0.108 0.194 0.035 0.054 0.000

A18 Appendix 3.4: Titanomagnetite analyses

5 0.167 12.806 2.892 76.952 0.443 2.686 0.109 0.177 0.030 0.134 0.204 6 0.114 12.664 3.066 77.152 0.476 2.609 0.052 0.087 0.037 0.141 0.072 Mean 0.157 12.462 3. 114 76 .816 0.410 2.792 0.084 0.178 0.028 0.113 0.094

Sample 93.46 T20/468102 9 1 0.072 11.065 3.446 72.767 0.383 3.050 0.000 0.269 0.000 0.182 0.262 2 0.075 10.459 3.384 72.334 0.338 2.966 0.000 0.000 0.007 0.074 0.116 3 0.178 11.169 3.404 74.536 0.417 2.907 0.010 0.012 0.005 0.267 0.007 4 0.181 11.537 3.388 72.604 0.456 2.800 0.01 1 0.117 0.005 0.147 0.177 5 0.143 11.61 1 3.285 71 .670 0.395 2.781 0.000 0.057 0.015 0.155 0.018 6 0.218 11.046 3.523 74.197 0.361 3.244 0.004 0.000 0.042 0.161 0.123 Mean 0.145 11.148 3.405 73.018 0.392 2.958 0.004 0.076 0.012 0.164 0.1 17

Sample 93.47 T20/468102 9 0.053 8.380 6.032 76. 190 0.31 1 3.750 0.097 0.085 0.023 1.160 0. 111 2 0.034 8. 196 6.071 77.164 0.186 4.054 0.079 0.096 0.023 1.170 0.100 3 0.126 8.819 5.357 78.235 0.248 3.654 0.1 10 0.079 0.007 0.392 0.247 4 0.363 8.763 5.616 77.391 0.389 3.893 0.090 0.342 0.000 0.856 0.055 5 0.108 8.650 5.425 79.506 0.289 4.139 0.128 0.164 0.036 0.473 0. 106 6 0.136 8.608 4.990 78.559 0.450 3.815 0.074 0.073 0.012 0.538 0.078 Mean 0.137 8.569 5.582 77.841 0.312 3.884 0.096 0.140 0.017 0.765 0.116

Sample 93.48 T20/468102 9 0.226 9.423 4.627 74.929 0.284 3.509 0.036 0.139 0.023 0.450 0.000 2 0.054 7. 177 3.467 81.476 0.785 2.751 0.077 0.218 0.053 0.118 0.007 3 0.139 9.226 4.507 76.258 0.245 3.305 0.008 0.000 0.027 0.134 0.106 4 0.192 9. 198 4.602 76.826 0.267 3.551 0.037 0.002 0.012 0.448 0.000 5 0. 134 8.969 4.477 76.863 0.301 3.488 0.079 0.013 0.01 7 0.209 0.060 Mean 0.149 8.799 4.336 77.270 0.376 3.321 0.047 0.074 0.026 0.272 0.035

Sample 93.49 T20/468102 9 1 0.062 8.734 4.358 81.060 0.387 3.553 0.046 0.165 0.000 0.255 0. 103 2 0. 129 8.644 4.528 77.981 0.403 3.737 0.049 0.235 0.004 0.143 0.026 3 0. 159 8.594 4.609 79.361 0.337 3.590 0.054 0.265 0.019 0.448 0.030 4 0.147 8.721 4.596 78.31 1 0.438 3.550 0.049 0.208 0.000 0.224 0.154 5 0.121 8.651 4.295 78.439 0.277 3.252 0.058 0.164 0.040 0.289 0.035 6 0.085 8.633 4.882 77.992 0.326 3.689 0.030 0.134 0.021 0.1 15 0. 113 Mean 0. 117 8.663 4.545 78.857 0.361 3.562 0.048 0.195 0.014 0.246 0.077

Sample 93.50 T20/468102 9 0.166 10.861 3.261 77.722 0.468 2.691 0.089 0.199 0.021 0.229 0.000 2 0. 127 13.272 2.684 78.850 0.622 2.412 0.070 0.209 0.006 0.097 0.022 3 0.042 13.014 2. 782 77.845 0.583 2.161 0.024 0.003 0.058 0.217 0.180 4 0. 124 12.987 2.881 77.854 0.382 2.617 0.070 0.187 0.030 0.081 0.029 5 0.036 10.829 3.122 80.272 0.493 2.912 0.021 0.000 0.054 0.025 0.073 6 0.021 9.374 4.379 81.128 0.287 3.065 0. 116 0.000 0.014 0.093 0.071 Mean 0.086 11.723 3.185 78.945 0.473 2.643 0.065 0.100 0.031 0.124 0.063

Sample 93.51 T20/468102 9 1 0.108 8.397 3.879 79.537 0.323 3.334 0.044 0.143 0.027 0.396 0.215 2 0.137 8.868 3.746 79.203 0.424 3.245 0.000 0.082 0.055 0.203 0.155 3 0.137 9.080 3.778 77.719 0.392 3.225 0.035 0.000 0.052 0.050 0.283 4 0.189 8.690 3.407 77.837 0.282 3.005 0.105 0.000 0.042 0.168 0.039 5 0.182 9.008 3.967 78.818 0.389 3.223 0.025 0.008 0.024 0.184 0.225 Mean 0.151 8.809 3.755 78.623 0.362 3.206 0.042 0.047 0.040 0.200 0.183

Sample 93.52 T20/468102 10 1 0.167 7.828 4.515 78.222 0.372 3.649 0.117 0.150 0.000 0.925 0.127 2 0.184 7.51 8 3.617 74.289 0.252 2.501 0.093 0.009 0.014 0.864 0.254 3 0.484 7.921 4.842 77.113 0.338 3.883 0.045 0.605 0.022 0.577 0.118 4 0.475 7.763 5.256 76.502 0.375 3.827 0.026 0.504 0.009 0.796 0.245 5 0.508 8.018 5.41 1 78.048 0.407 3. 730 0.091 0.723 0.002 0.570 0.229 6 0.505 7.738 5.140 77.826 0.308 3.864 0.071 0.490 0.014 0.556 0.174 Mean 0.387 7.798 4.797 77.000 0.342 3.576 0.074 0.414 0.010 0.715 0.191

Sample 93.53 T20/468102 10 0.138 9.475 4.974 76.666 0.298 2.400 0.048 0.000 0.007 0.375 0.129 2 0.153 8.516 5.051 74.663 0.303 3.393 0.051 0.002 0.007 1.050 0.091

Sample 93.54 T20/468102 10 1 0.151 9.475 3.535 71.564 0.305 3.41 1 0.208 0.002 0.013 0.568 0.114 2 0.147 10.259 3.579 71.035 0.332 3.125 0.070 0.168 0.024 0.740 0.098 3 0.072 9.810 3.549 71 .505 0.344 3.150 0.020 0.000 0.022 0.717 0.094 4 0.176 9.91 1 3.845 72.889 0.270 3.413 0.048 0.093 0.030 0.998 0.071 Mean 0.137 9.864 3.627 71 .748 0.313 3.275 0.087 0.066 0.022 0.756 0.094

Sample 93.55 T20/468 102 10 0.126 12.801 2.956 75.422 0.282 2.425 0.035 0.077 0.005 0.263 0.122 2 0.281 12.248 3.042 75.073 0.368 2.725 0.032 0.268 0.000 0.540 0.169 3 0. 189 10.243 3.336 76.203 0.222 3.016 0.025 0.251 0.000 0.477 0.104 Mean 0.199 11.764 3.1 11 75.566 0.291 2.722 0.031 0.199 0.002 0.427 0.132

Sample 93.57 T20/468102 10 0.134 10.197 3.348 74.069 0.294 2.857 0.045 0.000 0.030 0.459 0. 116 2 0.167 10.279 3.498 74. 154 0.322 3.181 0.036 0.113 0.019 0.373 0.000

A 19 Appendix 3.4: Titanomagnetite analyses

3 0.080 10.590 3.423 73.334 0.359 3. 192 0.002 0.181 0.075 0.337 0.029 4 0. 129 10.068 3.608 74.440 0.425 3.216 0.026 0.000 0.023 0.368 0.000 5 0.107 10.472 3.268 73.414 0.41 1 3.057 0.038 0.013 0.065 0.201 0.020 6 0.158 10.095 3.417 74. 138 0.348 2.965 0.027 0.081 0.027 0.497 0.000 Mean 0.129 10.284 3.427 73.925 0.360 3.078 0.029 0.065 0.040 0.373 0.028

Sample 93.58 T20/468102 10 0.069 12.096 3.033 74.382 0.344 2.382 0.054 0.336 0.060 0.162 0. 119 2 0.060 12.133 3. 123 73.764 0.344 2.422 0.045 0.162 0.016 0.092 0. 155 3 0.039 11.993 3.064 72.638 0.338 2.421 0.050 0.000 0.040 0.128 0.018 4 0.064 11.659 2.530 72.793 0.357 2. 180 0.073 0.000 0.062 0.091 0.064 5 0.141 12.279 2.801 74.147 0.348 1.849 0.056 0.092 0.035 0.218 0.004 6 0.401 12.300 2.993 73.298 0.379 2.588 0.075 0.001 0.045 0.000 0.031 Mean 0.129 12.077 2.924 73.504 0.352 2.307 0.059 0.099 0.043 0.115 0.065

Sample 93.61 T20/477112 14 0. 137 9.401 4.246 71 .746 0.351 3. 129 0.186 0.142 0.061 0.295 0. 134 2 0.151 9.277 3.965 73.577 0.417 2. 178 0.191 0.141 0.035 0.339 0.075 3 0.127 8.589 3.783 74.333 0.669 3.073 0.063 0.000 0.028 0.235 0.088 4 0.221 8.551 3.630 74.541 0.664 3. 127 0.059 0.201 0.067 0.300 0.119 5 0. 177 9.104 4.504 73.108 0.346 3.175 0.026 0.000 0.065 0.672 0.093 6 0. 112 8.514 5.068 73.333 0.425 3.094 0.029 0.128 0.055 0.412 0.000 Mean 0.154 8.906 4. 199 73.440 0.479 2.963 0.092 0.102 0.052 0.376 0.085

Sample 93.62 T20/4 771 12 14 1 0.195 8.637 4.734 71.337 0.284 3.716 0.251 0.075 0.017 2.231 0.165 2 0.122 8.682 4.793 72. 131 0.323 3.422 0.271 0.004 0.000 1.788 0.169 3 0.056 9.286 4.422 75.708 0.422 3.365 0.090 0. 173 0.041 0.488 0.188 4 0.055 9.084 4.398 75.206 0.260 3.642 0.074 0.222 0.020 0.440 0.108 5 0.100 9.060 4.386 76.958 0.472 3.095 0.108 0.126 0.077 0.389 0.154 6 0.040 8.973 4.250 76 .123 0.243 3.578 0.1 10 0. 105 0.093 0.394 0.146 Mean 0.095 8.954 4.497 74.577 0.334 3.470 0.151 0.118 0.041 0.955 0.155

Sample 93.63 T20/477112 14 1 0.194 13.461 2.468 73.919 0.428 2.540 0.047 0.013 0.01 1 0.249 0.098 2 0.162 13.895 2.673 73.194 0.362 2.644 0.024 0.1 12 0.000 0.234 0.084 3 0.177 13.097 2.464 71 .750 0.378 2.914 0.144 0.090 0.046 0.419 0.245 Mean 0.178 13.484 2.535 72.954 0.389 2.699 0.072 0.072 0.019 0.301 0.142

Sample 93.65 T20/4771 12 14 1 0.191 12.698 3.074 72.225 0.414 3.525 0.073 0.003 0.006 0.335 0.168 2 0.169 12.071 3.148 71 .726 0.397 3.342 0.040 0.010 0.026 0.221 0.119 3 0.299 12.677 3.242 71 .745 0.337 3.596 0.108 0.083 0.025 0.358 0.113 Mean 0.220 12.482 3. 155 71 .899 0.383 3.488 0.074 0.032 0.019 0.305 0.133

Sample 93.66 T20/477112 14 0.252 11.207 2.977 73.905 0.427 2.857 0.080 0.1 71 0.007 0.330 0.089 2 0.136 11.361 3. 185 73.615 0.583 3. 024 0.003 0.334 0.020 0.382 0.046 3 0.270 10.946 3.268 71 .518 0.326 2.896 0.009 0.193 0.035 0.523 0.01 1 4 0. 187 11.190 3.266 72.344 0.333 3. 068 0.000 0.248 0.007 0. 310 0.000 5 0.083 10.892 3.229 71.603 0.396 2.961 0.066 0.168 0.015 0.605 0.050 Mean 0.186 11.119 3.185 72.597 0.413 2.961 0.032 0.223 0.017 0.430 0.039

Sample 93.67 T20/477112 14 1 0.093 12.469 2.997 73.785 0.292 2.522 0.092 0.013 0.018 0.166 0.140 2 0.099 12.198 3.040 74.392 0.271 2.269 0.057 0.192 0.016 0.125 0.130 3 0. 134 11.603 3.065 74.805 0.343 2.486 0.075 0.060 0.008 0.000 0.245 4 0.097 9.284 4.334 72.928 0.221 3.543 0.096 0.005 0.021 0.125 0.149 5 0. 152 13.798 3.153 73.027 0.334 2.296 0.044 0.067 0.048 0.343 0.1 13 Mean 0. 115 11.870 3.318 73.787 0.292 2.623 0.073 0.067 0.022 0.152 0.155

Sample 93.68 T20/477112 14 1 0.108 9.353 3.877 75 .147 0.266 3.505 0.054 0.115 0.040 0.300 0.101 2 0.059 10.193 4.364 76. 134 0.344 3.328 0.072 0.009 0.006 0.145 0.106 3 0.128 10.222 4.100 76.229 0.366 3.313 0.075 0.170 0.040 0.222 0.149 4 0.185 10.294 4.419 75.427 0.380 2.899 0.088 0. 159 0.017 0.227 0.065 5 0.076 10.097 4. 194 76.650 0.287 2.986 0.049 0.005 0.001 0.384 0.124 Mean 0.111 10.032 4.191 75.917 0.329 3.206 0.068 0.092 0.021 0.256 0.109

Sample 93.69 T20/477112 14 1 0.123 9.878 3.940 73.899 0.467 3.492 0.053 0.000 0.037 0.285 0.080 2 0.101 9.796 3.863 71.916 0.358 3.470 0.017 0.095 0.033 0.492 0.167 3 0.086 9.792 3.888 73.193 0.415 3.522 0.060 0.053 0.038 0.221 0.178 4 0.244 10.136 4.187 73.747 0.527 3.697 0.025 0.285 0.037 0.385 0.120 5 0.046 10.117 3.666 72.404 0.286 3.262 0.023 0.004 0.034 0.188 0.040 Mean 0. 120 9.944 3.909 73.032 0.41 1 3.489 0.036 0.087 0.036 0.314 0.117

Sample 93.70 T20/477112 14 1 0.075 7.775 5. 170 76.721 0.350 3.762 0.020 0.000 0.018 1.240 0.149 2 0.038 6.628 5.792 76.720 0.390 3.632 0.000 0. 011 0.021 1.735 0.146 3 0.057 9.048 4.610 76.465 0.372 3.704 0.016 0.064 0.037 0.679 0. 171 4 0.088 8.662 4.638 74.720 0.362 3.648 0.054 0.049 0.022 1.008 0.144 5 0.137 8.275 5.554 75. 105 0.228 4.106 0.094 0.076 0.000 1.208 0.118 Mean 0.079 8.078 5.153 75.946 0.340 3.770 0.037 0.040 0.020 1.174 0.146

A20 Appendix 3.4: Titanomagnetite analyses

Sample 93.71 T20/4771 12 14 1 0.192 8.659 5.540 74.731 0.376 3. 133 0.032 0.050 0.001 0.655 0.147 2 0.248 9.062 4.349 75.617 0.414 3.127 0.075 0. 100 0.006 0.395 0. 109 3 0.100 8.862 5.916 76.227 0.376 3.398 0.045 0. 118 0.023 0.749 0.000 4 0.170 9.042 4.993 76.009 0.420 3.757 0.068 0.000 0.000 0.508 0.066 5 0.118 8. 714 4. 767 76.283 0.362 3.486 0.043 0.000 0.008 0.692 0.055 Mean 0.166 8.868 5. 113 75.773 0.390 3.380 0.053 0.054 0.008 0.600 0.075

Sample 93.72 T20/477112 14 0.146 8.429 4. 713 73.152 0.263 3.51 1 0.092 0.060 0.072 0.201 0.166 2 0.069 9.264 4.618 73.909 0.345 3.294 0.110 0.264 0.014 0.115 0.065 3 0.199 8.757 4.477 74.937 0.271 3.051 0.042 0.000 0.020 0.148 0.035 4 0.227 8.934 4.493 76.585 0.332 3.064 0.054 0. 182 0.002 0.123 0.015 5 0.076 9.326 4.301 77 .194 0.283 2.876 0.058 0.000 0.000 0.000 0.096 6 0.141 9.245 4.657 77.033 0.31 3 3. 113 0.073 0.1 12 0.000 0.153 0.050 7 0.066 9.099 4.233 77.102 0.491 3.334 0.060 0.067 0.000 0.083 0.000 Mean 0.132 9.008 4.499 75.702 0.328 3. 178 0.070 0.098 0.015 0.1 18 0.061

Sample 93.73 T20/477112 14 1 0.082 10.606 3.015 78.327 0.516 2.358 0.055 0.1 16 0.030 0.082 0.032 2 0. 116 10.290 3. 138 78.353 0.497 2.387 0.054 0.129 0.000 0.176 0.026 3 0.078 9.976 3. 198 79.005 0.927 2.702 0.007 0.047 0.044 0.358 0. 129 4 0.091 9.327 3.217 78.979 0.449 2.771 0.044 0.107 0.012 0.194 0.039 5 0.001 8.586 3. 787 78.540 0.392 3. 196 0.045 0.007 0.019 0.138 0.113 6 0.068 10.497 2.835 78.635 0. 765 2.559 0.059 0.170 0.000 0.229 0.142 7 0.074 10.315 2.881 79.208 0.498 2.448 0.020 0.21 1 0.024 0.118 0.073 Mean 0.073 9.942 3.153 78.721 0.578 2.631 0.041 0. 112 0.018 0. 185 0.079

Sample 93.74 T20/477112 14 1 0.1 12 7.572 4.908 74.600 0.272 3.664 0.036 0.005 0.000 0.905 0.231 2 0.105 7.796 5.016 74.661 0.346 3.406 0.064 0.000 0.020 0.787 0.085 3 0. 128 7.852 5.073 74.946 0.291 3.776 0.042 0.096 0.029 0.913 0.008 4 0.149 7.777 5.275 75.274 0.361 3.661 0.041 0.000 0.018 0.477 0.121 5 0.141 7.933 4.798 74.792 0.342 3.775 0.057 0.326 0.010 0.850 0.079 6 0.1 13 7.751 4.648 73.467 0.419 3.741 0.044 0.21 7 0.042 0.696 0.079 Mean 0.125 7.780 4.953 74.623 0.339 3.671 0.047 0.107 0.020 0.771 0.101

Sample 93.76 T20/477112 14 1 0.146 8.393 4.967 73.745 0.245 3.470 0.208 0.050 0.022 1.754 0.153 2 0.01 3 6.965 6.886 70.919 0.337 4.234 0.110 0.003 0.000 2.954 0.038 3 0.145 10.130 4.423 73.979 0.344 3.244 0.039 0.140 0.000 0.554 0.138 4 0.083 9.007 5.602 73.345 0.255 3.366 0.040 0.012 0.000 0.387 0.102 5 0. 102 9.054 6.095 73.659 0.416 3.606 0.091 0.121 0.000 0.278 0.133 6 0.191 8.215 6.048 72.452 0.318 4.086 0.050 0.007 0.000 0.644 0.075 Mean 0.113 8.627 5.670 73.017 0.319 3.668 0.090 0.056 0.004 1.095 0.107

Sample 93.77 T20/477112 14 1 0.136 9. 187 3.068 75.764 0.390 2.413 0.061 0.058 0.025 0.321 0.121 2 0.186 9.571 3.434 73.568 0.358 2.476 0.096 0.073 0.013 0.426 0.077 3 0.261 9.196 3.833 75.294 0.292 3.094 0.070 0.105 0.034 0.033 0.127 4 0.168 9.752 3.518 74.560 0.265 2.661 0.065 0.327 0.017 0.472 0. 121 5 0.217 9.468 4.692 73.719 0.41 8 3.045 0.055 0.365 0.027 1.045 0.000 6 0.140 9.616 4.51 1 74.002 0.325 2.759 0.051 0.134 0.037 0.349 0.098 Mean 0.185 9.465 3.843 74.485 0.341 2.741 0.066 0.177 0.026 0.441 0.091

Sample 93.78 T20/4771 12 14 1 0.152 11.492 3.148 71.313 0.290 2.697 0.060 0.1 13 0.023 0.382 0.100 2 0.136 12.317 3. 107 72.652 0.364 2.558 0.042 0.1 13 0.000 0.063 0.004 3 0.099 12.045 2.864 72.389 0.457 2.632 0.057 0.000 0.033 0.303 0.059 4 0. 102 12.590 2.928 71 .496 0.465 2.508 0.029 0.000 0.032 0.063 0.055 5 0. 158 12.427 2.867 75.373 0.341 1.849 0.066 0.000 0.016 0.334 0.009 6 0.074 12.169 2.678 71.797 0.351 2.028 0.012 0.063 0.000 0.01 1 0.167 Mean 0. 120 12.173 2.932 72.503 0.378 2.379 0.044 0.048 0.017 0.193 0.066

Sample 93.79 T20/477112 14 0.020 10.905 3.931 72.338 0.360 2.622 0.000 0.000 0.01 1 0.182 0.141 2 0.004 9.655 3.530 72.257 0.316 2.857 0.000 0.068 0.017 0.201 0.203 3 0.124 9.448 3.525 74.830 0.426 2.830 0.019 0.121 0.026 0.338 0.003 4 0.111 9.745 3.404 73.883 0.275 3.143 0.049 0.214 0.019 0.128 0.000 5 0.061 9.484 4. 105 75.692 0.248 3.308 0.024 0.193 0.016 0.163 0.000 6 0.003 9.220 3.866 75.772 0.541 3.313 0.046 0.122 0.01 3 0.371 0.057 7 0.019 9.300 4.098 74.945 0.413 3.150 0.040 0.312 0.040 0.326 0.000 Mean 0.049 9.680 3.780 74.245 0.368 3.032 0.025 0.147 0.020 0.244 0.058

Sample 93.80 T20/477112 14 0.127 8.383 3.774 71 .401 0.290 3.1 12 0.054 0.203 0.018 0.673 0.083 2 0.223 8.161 3.976 72.201 0.322 3.243 0.080 0.210 0.016 0.805 0.097 3 0.148 8.818 4.677 73.308 0.287 3.428 0.050 0.078 0.029 0.678 0.085 4 0.131 8.843 4.676 72.237 0.31 2 3.397 0.046 0.096 0.033 0.678 0.167 5 0.140 8.539 4.756 72.036 0.281 3.349 0.047 0.000 0.018 0.734 0.045 Mean 0.154 8.549 4.372 72.237 0.298 3.306 0.055 0. 117 0.023 0.714 0.095

A 21 Appendix 3.4: Titanomagnetite analyses

Sample 93.81 T20/4771 12 14 1 0.088 8.889 4.824 73.095 0.305 3.245 0.141 0.130 0.024 0.677 0. 114 2 0.206 8.894 3.850 72.533 0.568 3.190 0.209 0.000 0.034 0.641 0.234 3 0.351 8.965 4.730 72.499 0.265 3.056 0.103 0.330 0.048 0.814 0.169 4 0.186 8.754 4.868 72.985 0.431 3.326 0.126 0.228 0.01 1 0.420 0. 134 5 0.289 8.91 7 5.036 75. 734 0.259 3.400 0.075 0.202 0.008 0.734 0.140 6 0.149 8.678 4.603 74.009 0.340 3.599 0.045 0.189 0.000 0.831 0.095 Mean 0.212 8.850 4.652 73.476 0.361 3.303 0.1 17 0.180 0.021 0.686 0.148

Sample 93.85 T20/477112 14 1.118 9.482 5.579 77.683 0.442 3.234 0.026 0.000 0.053 0.613 0.100 2 0.153 8.262 5.687 75. 843 0.368 3.313 0.065 0.049 0.005 0.459 0.155 3 0.091 7.844 5.597 77.374 0.362 3.270 0.065 0.121 0.015 0.123 0.076 4 0.108 8.121 5.347 74.795 0.239 3.242 0.055 0.124 0.013 0.558 0.193 5 0. 130 8.253 5.457 76.605 0.232 3.189 0.074 0.000 0.000 0.300 0.122 6 0.050 8.072 5.369 76.567 0.450 3.1 13 0.030 0.000 0. 015 0.321 0.170 7 0.149 7.689 5.206 75.524 0.185 2.747 0.032 0.000 0.044 0.493 0.162 Mean 0.257 8.246 5.463 76.342 0.325 3. 158 0.050 0.042 0.021 0.410 0.140

Sample 93.86 T20/477112 14 0.291 7.812 5.366 75.436 0.380 3.383 0.044 0.083 0.020 0.320 0.031 2 0.063 7.836 5.417 75. 190 0.350 3.273 0.051 0.000 0.024 0.227 0.004 3 0.104 8.156 5.358 76.492 0.250 3.480 0.047 0.009 0.013 0.306 0.000 4 0.070 7.576 5.705 76.447 0.296 3.087 0.039 0.000 0.035 0.236 0.046 5 0.121 7.835 5. 249 75.392 0.361 3.283 0.066 0.000 0.039 0.196 0.042 6 0.224 8.144 5.159 75.055 0.379 3.338 0.057 0.000 0.027 0.129 0.000 Mean 0.146 7.893 5.376 75.669 0.336 3.307 0.051 0.015 0.026 0.236 0.021

Sample 93.87 T20/4771 12 14 0.166 8.675 3.260 77.240 0.207 2.251 0.083 0.000 0.065 0.189 0.128 2 0.148 8.551 3.322 75.435 0.21 7 2.258 0.058 0.005 0.020 0.196 0.106 3 0.185 8.566 3.431 76.824 0.395 2.290 0.086 0.001 0.028 0.549 0.187 4 0.145 8.065 3.508 77.470 0.454 2.785 0.124 0.103 0.018 0.293 0.090 5 0.126 8.434 3.374 75.478 0.346 2.952 0.051 0.155 0.024 0.197 0.077 6 0. 102 8.357 3.367 75.870 0.261 2.629 0.100 0.000 0.041 0.068 0.088 Mean 0.145 8.441 3.377 76.386 0.313 2.528 0.084 0.044 0.033 0.249 0.113

Sample 93.88 T20/477112 14 1 0.085 6.554 2.674 77.581 0.209 1.753 0.097 0.136 0.017 0.975 0.290 2 0.147 7.438 2.997 78.837 0.366 1.912 0.070 0.000 0.041 0.337 0.110

Sample 93.89 T20/4771 12 14 1 0.020 21 .450 0.706 68.764 0.225 1.890 0.060 0.057 0.000 0.068 0.000 2 0.013 21 .634 0.793 67.725 0.263 1.905 0.101 0.221 0.013 0.185 0.000 3 0.045 21.590 0.714 68.530 0.137 1.971 0.123 0.213 0.010 0.137 0.046 4 0.026 21 .888 0.758 69.025 0.133 1.912 0.147 0.068 0.025 0.158 0.048 5 0.023 21 .368 0. 702 67.897 0.258 1.889 0.125 0.221 0.054 0.1 11 0.009 6 0.068 21.547 0.788 67.996 0.225 1.928 0.098 0.167 0.014 0.215 0.012 Mean 0.033 21 .580 0.744 68.323 0.207 1.916 0.109 0.158 0.019 0.146 0.019

Sample 93.91 T20/4771 12 14 1 0.198 11.078 3.479 75.877 0.383 2.953 0.037 0.121 0.010 0.000 0.097 2 0.067 11.006 3.509 75.397 0.352 2. 864 0.044 0.270 0.012 0.286 0.129

Sample 94.2 T20/455088 1 0.117 12.148 2.849 75.248 0.290 2.849 0.066 0.222 0.036 0.289 0.108 2 0.1 16 11.443 3.162 74.930 0.449 2.977 0.097 0.281 0.032 0.221 0.009 3 0.387 11.050 3.804 73.801 0.341 2.949 0.069 0.288 0.026 0.063 0.092 4 0.233 11.356 3.484 74.641 0.409 3.327 0.107 0.460 0.064 0.389 0.129 5 0.240 11.057 3.629 74.812 0.325 3.260 0.101 0.158 0.073 0.279 0.261 6 0.177 12.090 3.136 75.044 0.414 2.961 0.062 0.240 0.044 0.367 0.053 Mean 0.212 11.524 3.344 74.746 0.371 3.054 0.084 0.275 0.046 0.268 0.109

Sample 94.4 T20/469102 9 0.112 12.691 2.838 72.873 0.401 2.826 0.071 0.003 0.006 0.695 0.159 2 0.134 12.725 2.749 72.201 0.325 2.874 0.085 0.000 0.038 0.778 0.157 3 0.164 12.881 2.857 72.826 0.451 2.854 0.061 0.012 0.024 0.967 0.113 4 0.139 12.488 2.772 73.833 0.358 2.553 0.062 0.290 0.014 2. 159 0.156 5 0.1 11 12.685 2.690 72.670 0.363 2.993 0.049 0.000 0.013 0.888 0.165 Mean 0.132 12.694 2.781 72.881 0.380 2.820 0.066 0.061 0.019 1.097 0.150

Sample 94.13 T20/496157 51 0.341 9.784 2.863 71 .946 0.368 2.427 0.177 0.1 19 0.040 0.359 0.082 2 0.197 9.528 3.609 75.460 0.417 3.079 0.056 0.348 0.055 0.496 0.170 3 0.300 9.536 3.534 75.653 0.313 2.974 0.086 0.412 0.006 0.408 0.109 4 0.370 7.653 4.963 74.810 0.545 3.597 0.032 0.504 0.081 0.260 0.150 5 0.240 7.828 5.063 75. 869 0.360 3.467 0.045 0.371 0.012 0.396 0. 187 6 0.398 7.838 4.721 75.828 0.347 3.449 0.113 0.249 0.000 0.474 0. 147 Mean 0.308 8.695 4.126 74.928 0.392 3.166 0.085 0.334 0.032 0.399 0.141

Sample 94. 15 T20/463097 1 0.131 6.964 4.255 79.809 0.415 3.231 0.030 0.001 0.002 0.139 0.382 2 0.075 7.215 4.51 2 80.209 0.401 3.068 0.058 0.000 0.000 0.123 0.318 3 0.244 7.469 4.21 1 78.315 0. 366 2.824 0.027 0.075 0.002 0.000 0.262 4 0.171 7.169 4.420 80.409 0.273 3.065 0.027 0.102 0.007 0.014 0.374

A22 Appendix 3.4: Titanomagnetite analyses

5 0.129 7.052 4.271 77.857 0.553 2.862 0.022 0.006 0.014 0.142 0.119 6 0.1 14 6.926 4.440 78.167 0.329 2.857 0.048 0.000 0.023 0.050 0.166

Mean 0.144 7. 133 4.352 79.128 0.390 2.985 0.035 0.031 0.008 0.078 0.270

Sample 94. 16 T20/463097 0.071 8.547 3.438 80.490 0.329 2.604 0.059 0.000 0.048 0.066 0.006 2 0.061 9. 143 2.878 79.208 0.376 2.456 0.017 0.490 0.010 0.210 0.090 3 0. 154 9.500 2.728 78.471 0.381 2.218 0.026 0.008 0.022 0.058 0.110 4 0. 178 8.384 3.399 78.382 0.319 2.791 0.067 0.138 0.009 0.092 0.045 5 0.070 8.276 3.434 77.815 0.299 2.752 0.099 0.059 0.023 0.019 0.042 6 0.230 9.220 3.449 76.361 0.269 2.881 0.058 0.408 0.038 0.000 0.000

Mean 0.127 8.845 3.221 78.455 0.329 2.617 0.054 0.184 0.025 0.074 0.049

Sample 94. 17 T20/463097 1 0. 178 8.486 3.285 77.041 0.412 3.068 0.189 0.047 0.000 0.236 0.122 2 0. 124 8.730 3.262 76.470 0.436 3.034 0.158 0.075 0.000 0.100 0.039 3 0.174 9.142 2.913 76.429 0.325 3.027 0.076 0.221 0.029 0.163 0. 153 4 0.172 8.585 3.169 79.089 0.368 3.033 0.121 0.000 0.01 1 0.106 0.085 5 0.406 9.375 3.170 78.498 0.398 3.075 0.139 0.076 0.009 0.125 0.057

Mean 0.21 1 8.864 3.160 77.505 0.388 3.047 0.137 0.084 0.010 0.146 0.091

Sample 94. 18 T20/498184 53 0.130 8.886 2.836 78.639 0.446 2. 182 0.078 0.051 0.007 0.261 0.184 2 0.147 9.101 2.853 78. 138 0.421 2.300 0.082 0.084 0.021 0.409 0.184 3 0.257 9.364 2.685 77.943 0.454 2.390 0. 107 0.165 0.017 0.346 0.213 4 0.164 8.967 2.437 82.107 0.399 2.251 0.100 0.150 0.071 0.341 0.235

Mean 0.175 9.080 2.703 79.207 0.430 2.281 0.092 0.113 0.029 0.339 0.204

Sample 94.22 T1 9/48121 1 55 1 0.639 6.614 2.936 78.826 0.410 2.473 0.1 10 0.876 0.049 0.956 0.233 2 0.361 6.551 2.670 80.201 0.456 1.944 0.067 0.426 0.032 0.898 0.241 3 0.358 4.709 3.359 79.385 0.458 1.488 0.080 0.493 0.045 0.718 0.219

Mean 0.453 5.958 2.988 79.471 0.441 1.968 0.086 0.598 0.042 0.857 0.231

Sample 94.31 T1 9/495220 59 1 0.155 5.886 3.544 79.209 0.561 2.190 0.054 0.252 0.096 0.183 0.138 2 0.205 6.231 3.885 79.874 0.480 2.594 0.046 0.393 0.018 0.332 0.214 3 1.399 6.227 3.444 76.714 0.464 2.793 0.108 0.196 0.01 1 0.327 0.230 4 0.109 6.561 3.376 78.639 0.887 2.069 0.050 0.013 0.302 0.386 0.362 5 0.568 6.422 3.408 79.459 0.402 1.885 0.268 0.485 0.021 0.220 0.135 6 0. 175 6.538 3.680 79.380 0.340 2.280 0.059 0.448 0.015 0.177 0.195

Mean 0.435 6.31 1 3.556 78.879 0.522 2.302 0.098 0.298 0.077 0.271 0.212

Sample 94.45 T1 9/496238 66 0.063 7.093 2.357 82.463 0.673 1. 789 0.137 0.162 0.022 0.000 0.115 2 0.265 7.940 1.246 79.189 0.664 0.736 0.400 0.337 0.045 0.1 11 0.047 3 0.362 6.830 2.053 80.866 0.574 1.478 0.137 0.447 0.060 0.036 0.137 4 0.717 6.262 2.360 84. 179 0.709 2.039 0.145 0.740 0.034 0.116 0.123

Mean 0.352 7.031 2.004 81 .674 0.655 1.511 0.205 0.422 0.040 0.066 0.106

Sample 94.79 T19/364544 73 1 0.122 13.233 1.971 75.269 0.233 1.118 0.089 0.000 0.026 0.376 0. 119 2 0.092 12.381 2.977 77.863 0.438 2.588 0.056 0. 119 0.024 0.187 0.123 3 0.425 13.032 2.282 77.867 1.022 1.452 0.098 0.486 0.081 0.395 0.169 4 0.475 12.603 2.335 77.414 1.031 1.459 0.166 0.574 0.059 0.291 0.218

Mean 0.279 12.812 2.391 77. 103 0.681 1.654 0. 102 0.295 0.048 0.312 0.157

A23 Appendix 3.5: Other phases

3.5 Analyses of other phases

All analyses are in a raw form (not normalised).

3.5.1 Chromite

Sample Location Section Analysis Si02 Ti02 Al203 FeO MnO MgO CaO Na20 K20 Cr203 NiO

Sample 93.28 T20/465099 47 0.171 3.434 13.774 49.795 0.179 7.638 0.030 0.064 0.057 19.472 0.313

Sample 93.31 T20/466096 5 1 0.669 1.415 10.371 46.440 0.295 7.401 0.096 0.314 0.046 27.857 0.247 2 0.000 1.155 9.262 42.565 0.339 6.343 0.083 0.013 0.020 31 .112 0.134 3 0.181 3.750 10.425 50.921 0.328 6.230 0.057 0.006 0.006 18.743 0.282 Mean 0.283 2.107 10.019 46.642 0.321 6.658 0.079 0. 111 0.024 25.904 0.221

Sample 93.56 T20/468102 10 1 0.085 5.317 7.478 67.042 0.486 3.951 0.082 0.188 0.026 7.972 0.076 2 0. 162 5.579 7.548 67.921 0.267 4.267 0.109 0.186 0.022 7.756 0.040

Sample 93.63 T20/477112 14 0.481 4.758 7.568 62.560 0.353 4.864 0.073 0.316 0.063 12.154 0.283

3.5.2 Clinopyroxene Si02 Ti02 Al203 FeO MnO MgO CaO Na20 K20

Sample 93.26 T20/464098 46 51 .789 0. 156 1.273 4.737 0.042 18.744 20.089 0.225 0.046 Sample 93.31 T20/466096 5 50.273 0.399 2.459 8.477 0.395 15.281 20.629 0.390 0.026 Sample 93.38 T20/469100 7 49.034 0.544 3.454 9.680 0.550 14.289 20.098 0.388 0.235 2 49.818 0.577 2.331 9.350 0.552 15.603 19.059 0.361 0.01 1 Sample 93.39 46.641 0.574 2.51 1 11.376 0.466 13.052 18.1 19 0.381 0.026 Sample 93.54 T20/468102 10 53.070 0. 133 1.558 3.871 0.084 19.295 19.650 0.316 0.043 Sample 93.65 T20/477112 14 49.828 0.684 3.655 9.689 0.186 14.372 19.633 0.405 0.015 2 50 .154 0.513 2.070 10.085 0.318 14.796 19.494 0.401 0.032 Sample 93.72 52.596 0.666 3.226 10.453 0.31 2 15.944 18.81 1 0.268 0.024 Sample 93.75 51 .637 0.560 3. 793 9.191 0.277 15.501 20. 101 0.453 0.038 Sample 93.76 54.189 0.354 2.357 9.436 0.200 16.015 19.500 0.007 0.000 Sample 93.85 47.689 0.495 2.029 9.653 0.298 15.002 19.828 0.298 0.040 Sample 93.89 51 .322 0.538 2.316 10.588 0.255 16.235 17.580 0.358 0.024 Sample 93.90 45.908 0.553 0.280 12.020 0.378 13.587 17.229 0.000 0.065 2 47.857 0.487 2.071 10.000 0.000 14.400 17.842 0.397 0.077 Sample 93.91 50.916 0.419 1.943 9.168 0.314 14.616 20.208 0.471 0.036 Sample 94.31 T1 9/495220 59 50.541 0.374 2.076 8.763 0.436 13.570 20.474 0.481 0.052 Sample 94.43 T1 9/496238 66 50.433 0.498 1.891 10.152 0.549 15.377 19.670 0.267 0.022 Sample 94.45 1 51 .436 0.232 1.285 9.643 0.748 15.161 20.327 0.377 0.028 2 48.899 0.375 2.013 8.755 0.656 14.120 20.432 0.355 0.027

3.5.3 Orthopyroxene Si02 Ti02 Al203 FeO MnO MgO CaO Na20 K20

Sample 93.39 T20/469100 7 51 .508 0. 183 1.603 21.383 0.697 22.808 1.298 0.012 0.021 Sample 93.48 T20/468102 9 51 .683 0.271 1.597 17.540 0.595 25.643 1.614 0.093 0.042 Sample 93.50 51 .226 0.278 0.872 21.038 0.403 24.520 1.833 0.000 0.000 Sample 93.51 52.370 0.379 1.130 18.807 0.592 25.651 1.365 0.002 0.023 Sample 93.52 T20/468 102 10 1 52.755 0.245 1.140 19.186 0.485 24.091 1.692 0.044 0.010 2 52.534 0.309 1.453 16.690 0.469 26.030 1.576 0.007 0.016 Sample 93.54 51.392 0.573 4.821 12.719 0.263 28.184 1.264 0.037 0.008 Sample 93.63 T20/477112 14 49.565 0.246 2.126 14.528 0.278 25.495 1.133 0.047 0.087 Sample 93.69 45.959 1.742 1.786 24.686 0.457 23.476 1.477 0.000 0.027 Sample 93.70 51 .294 0.262 2.018 17.403 0.741 25.189 1.463 0.278 0.038 Sample 93.71 1 52.624 0.257 1.380 19.069 0.534 24.717 1.780 0.185 0.059 2 51 .905 0.334 2.589 18.476 0.549 25.249 1.629 0.092 0.025 3 51 .331 0. 170 2.109 18.249 0.448 23.562 1.538 0.048 0.000 Sample 93.73 1 54.189 0.313 1.406 20.280 0.508 23.715 1.532 0.107 0.068 2 51 .644 0.297 1.719 19.747 0.575 24.244 1.676 0.055 0.010 Sample 93.75 1 50.767 0.394 1.268 25.532 0.721 19.921 1.495 0.042 0.019 2 53.634 0.188 1.930 18.272 0.551 25.270 1.776 0.168 0.042 3 53.060 0.21 1 1.539 18.754 0.442 24.826 1.795 0.039 0.027 4 51 .524 0.261 1.416 21 .644 0.516 22.625 1.652 0.376 0.040 5 52.730 0.252 1.043 17.468 0.377 23.050 1.750 0.137 0.051 6 50.634 0.310 2.170 19.01 1 0.359 23.186 1.989 0.085 0.064 Sample 93.82 48.649 0.132 1.037 13.96 0.421 27.94 0. 731 0.092 0.035 Sample 93.89 52.333 0.302 1.868 19.39 0.522 23.72 1.622 0.099 0.069 Sample 94. 15 T20/463097 53.447 0.306 1.410 22.357 0.795 22.656 1.440 0.139 0.017

A24 Appendix 3.5: Other phases

Sample 94.43 T19/496238 66 52.398 0.269 1.212 19.297 0.587 24.961 1.596 0.069 0.039 2 52.792 0.241 1.353 23.254 0.873 20.247 1.553 0.077 0.000 3 53.534 0.169 0.772 18.897 0.708 25.121 1.439 0.033 0.020 4 54.651 0.171 0.738 14.354 0.589 27.954 1.471 0.069 0.009 5 51 .762 0.314 1.014 19.404 0.662 24.651 1.567 0.000 0.021 Sample 94.45 52.439 0.161 0.673 20.360 1.180 24.009 1.038 0.000 0.032 21 53.364 0.227 0.684 18.928 0.979 25.272 0.988 0.080 0.019 Sample 94.46 1 52.056 0.237 0.992 23.166 0.628 21 .032 2.829 0.080 0.000 2 53.135 0.191 2.994 12.737 0.224 28.448 1.892 0.054 0.054 3 48.358 0. 172 0.928 25.58 0.866 20.5 0.961 0.165 0.082 4 52.023 0. 156 1.164 20.51 1 0.765 23.754 1.354 0.093 0.067 5 52. 156 0.295 0. 765 20.688 0.936 23.427 1.712 0.000 0.050 Sample 94.79 T1 9/364544 73 52.936 0.792 7.267 16.824 0.353 14.912 2.900 1.283 1.732

3.5.4 Andesitic glass Si02 Ti02 Al 203 FeO MgO CaO Na20 K20

Sample 93.25 T20/464098 46 68.005 2.367 12.123 6.151 1.561 2.487 1.789 2.635 2 64.833 0.913 15.185 6.178 1.392 4.759 1.885 2.082 3 69.299 0.735 14.264 4.016 0.526 2.791 2.059 2.964 4 67. 151 0.767 14.420 4.614 1.325 4.034 4.492 1.621

Sample 93.26 T20/464098 46 60.428 0.941 19.278 4.268 1.635 7.356 4.306 2.21 7 2 60.551 0.977 18.865 3.938 2.077 7.103 4.212 2. 105 3 63.918 1.258 15.276 5.341 2.778 5.504 3.896 2.766 4 60.952 0.886 18.753 4.448 1.873 6.723 4.303 2. 135 5 63.651 1.205 15.720 5.624 2.320 5.659 4.045 2.688 6 65.023 1.122 14.108 5.501 2.319 5.548 3.852 3. 133 7 64.071 1.023 15.507 5.080 2.553 5.066 4.264 2.999 8 64.292 1.303 15.091 5.319 2.794 5.106 4.069 2.945 9 63.909 1.174 15.255 5.748 2.541 5.412 3.855 2.817 10 63.799 1.283 15.058 5.865 2.824 5.593 3.919 2.889

Sample 93.39 T20/469100 7 63.050 1.644 13.341 11.034 2.588 4.31 1 2.886 0.268 2 62.504 1.197 14.452 10.150 2.470 4. 102 3.025 1.383 3 63.205 1.845 13.987 10.955 2.251 3.204 2.204 1.642

Sample 93.72 T20/477112 14 62.932 0.781 16.264 6.163 1.741 5.380 3.166 1.531 2 61 .996 1.077 15.804 4.865 2.220 5.877 4.169 2.743 3 60.875 1.105 14.777 5.288 3.013 5.293 3.736 2.825

Sample 93.84 T20/477112 14 59.037 1.331 15.266 7. 112 2.661 5.629 3.055 1.975 2 58.355 1.062 16.482 5.903 2.548 5.639 3.617 1.768 3 56.410 0.960 16.051 6.870 2.840 5.458 3.639 1.898 4 56.791 0.940 15.904 7.238 2.923 5.935 3.699 1.776

Sample 93.89 T20/4771 12 14 1 63.985 1.024 15.363 7. 188 1.758 5.766 4.372 1.919 2 63.135 0.999 14.794 6.524 1.654 5.118 4.136 2. 118 3 62.703 0.967 14.808 6.177 1.757 4.805 4.283 2.056 4 63.549 0.819 15.065 6.797 1.796 4.999 4.326 2.221 5 62.287 1.123 14.819 6.404 1.748 5.042 3.906 2. 145

Sample 93.91 T20/477 112 14 1 62.077 1.274 13.453 5.295 1.613 4.679 3.705 2.438 2 62.485 1.130 13.950 6.390 2.221 5.185 3.578 2.444 3 61.695 1.127 15.200 6.059 1.777 5.760 4.030 2.051 4 64.966 1.350 14.266 6.492 2.049 5.263 4.000 2.391 5 63.471 1.233 14.760 6.260 2.097 5.480 3.707 2.407 6 63.785 1.242 13.932 6.278 1.905 5.215 3.877 2.526 7 64.503 1.126 13.992 6.329 1.960 5.091 3.734 2.396 8 62.677 1.126 14.142 6.319 2.069 5.041 3.654 2.416

Sample 94.46 T1 9/496238 66 1 65.793 0.630 14.819 3.910 1.418 3.947 3.895 2.509 2 63.858 0.968 15.335 5.788 1.788 4.888 3.994 2.073 3 64.326 1.043 15.060 6.358 1.521 4.866 3.727 2.558

Sample 94.79 T19/364544 73 1 64.649 1.213 12.073 6.969 2.548 2.883 2.312 2.803 2 62.324 1.684 13.024 6.584 2.262 2.954 2.540 2. 777 3 63.709 1.848 12.574 5.984 2.616 2.984 2.154 2.978 4 63.537 1.174 12.250 6.950 2.567 3.514 2.652 3.060

A25 Appendix 3.5: Other phases

3.5.5 Plagioclase Si02 Ti02 Al203 FeO MgO CaO Na20 K20

Sample 93.25 T20/464098 46 52.067 0.036 27.982 0.423 0.075 11.614 4.596 0.240 2 56.978 0. 133 27.224 0.833 0.129 10.976 3.778 0.678 3 52.744 0.078 29.174 0.440 0.073 11.909 4.781 0.216

Sample 93.80 T20/4771 12 14 49.899 0.039 30.616 0.748 0.057 14.143 3.888 0.117 21 57.891 0.358 23.944 1.102 0.321 9.998 3.145 0.684 3 48.098 0. 113 31 .247 0.608 0.131 15.425 2.954 0.068 51.819 0.173 25.598 1.498 0.091 10.687 5.716 0.404 4

Sample 93.89 T20/4771 12 14 1 46.220 0.077 30.989 0.655 0.081 16.973 2.416 0.067 2 53.316 0.060 28.289 0.619 0.221 10.871 4.745 0.320 3 52.697 0.088 29.794 0.743 0.101 13.221 3.194 0.293 4 52.165 0. 131 29.899 0.672 0.195 13.951 3.024 0.247 5 54.062 0.050 28.227 0.635 0.100 11.564 4.967 0.303

Sample 93.91 T20/4771 12 14 49.939 0.000 27.051 0.612 0.000 11.360 4.706 0.116 2 55.203 0. 121 27.331 0.550 0.065 10.097 4.777 0.345 3 52.148 0.000 30.480 0.281 0.009 13.476 4.351 0.148

3.5.6 llmenite Si02 Ti02 Al203 FeO MnO MgO CaO Na20 K20 Cr203 NiO

Sample 93.25 T20/464098 46 0.086 44.015 0.285 48.396 0.370 4.032 0.076 0.361 0.015 0.777 0.148 2 0 44.006 0.216 48.710 0.423 3.917 0.067 0.000 0.014 0.083 0.14 3 0.117 46.701 0.365 44.227 0.331 6.530 0.082 0.186 0.039 0.054 0. 156 4 0.015 43.481 0.428 48.857 0.438 4.247 0.079 0.005 0.008 0. 145 0.114 5 0.052 44.531 0.270 48.483 0.359 4. 140 0.061 0.087 0.040 0.062 0.149

Sample 93.51 T20/468102 9 0.186 36.568 0.590 55.435 0.254 2. 705 0. 114 0.005 0.013 0.068 0.049

Sample 93.88 T20/477112 14 0.702 28.867 0.690 58.933 0.101 1.603 0.377 0.010 0.020 0.155 0.048 2 0. 106 28.968 0.536 58.457 0.134 1.459 0.160 0.080 0.026 0.243 0.068

Sample 94.4 T20/468102 9 1 0.081 42.528 0.335 47.238 0.227 4. 141 0.280 0.074 0.023 0.31 1 0.265 2 0.000 46.474 0.100 46.924 2.071 0.935 0.073 0.097 0.029 0.262 0.227

Sample 94.22 T1 9/48121 1 55 1 0.262 28.556 0.633 60.858 0.234 1.858 0.172 0.526 0.034 0.233 0.117 2 0.1 10 32.024 0.652 58.834 0.182 1.952 0.124 0.081 0.014 0.177 0.098

Sample 94.41 T1 9/49 1239 1 0.071 47.027 0.227 46.423 0.474 3.342 0.104 0.006 0.025 0.027 0.008 2 0.065 46.263 0. 188 45.401 0.428 3.348 0.076 0.000 0.027 0.249 0 3 0.150 45.547 0.280 45. 190 0.459 3.366 0.053 0.196 0.044 0.033 0.025 4 0.065 46.292 0.269 46.690 0.493 3.380 0.050 0.086 0.020 0. 168 0.057 5 0.018 45.955 0.294 45.067 0.467 3.525 0.085 0.000 0.023 0.064 0

Sample 94.46 T19/496238 66 1 0.124 43.404 0.523 46.330 0.839 2.353 0.203 0.424 0.111 0.097 0.02 2 0.194 42.194 0.270 46.628 0.321 2.772 0.077 0.125 0.044 0.095 0.076 3 0.151 44. 127 0.401 51 .117 0.385 2.624 0.416 0.129 0.024 0.133 0.109 4 0.130 45.452 0.207 49.338 0.501 2.689 0.139 0.092 0.049 0.284 0.117

3.5.7 Spinel Si02 Ti02 Al203 FeO MnO MgO CaO Na20 K20 Cr203 NiO

Sample 93.53 T20/468102 10 0.000 1.167 51 .245 34 .108 0.174 12.404 0.020 0.000 0.000 0.100 0.049 2 0.000 0.975 51 .657 33.825 0.313 12.608 0.01 1 0.000 0.004 0.232 0 Sample 93.83 T20/477112 14 1.361 0.791 42. 720 39.567 0.242 9.305 0.224 0.147 0. 103 0.056 0.049

A26 APPENDIX 4: SELECTED FIELD SECTION DESCRIPTIONS

Of the > 120 sections described during the course of this study the following 60 are listed in order to characterise the stratigraphy within each stream or sector of the ring plains studied. The sections from which samples were taken fo r further analysis are also listed. The section numbers used are those assigned during the original field work and are consistent with the numbering system used throughout the thesis. The following indicates the section numbers describing each stream sector:

Upper Waikato Stream 3, 4, 5, 7, 9, 10, 14, 45, 46, 47 Tangatu Stream 11 Wharepu Stream 17, 18 Te Piripiri Stream 22, 24, 26, 30 Mangatoetoenui Stream 32, 34, 35, 37 Ohinepango Stream 38, 41, 42 Waihohonu Stream 40, 44, 53 Oturere Stream 54, 56, 81 Makahikatoa Stream 58, 59, 60 Mangatawai Stream 64, 66 Mangamate Stream 68 Puketarata Stream 91 Mangahouhounui Stream 93 Tauhurangi Stream 82, 83, 88, 89 Tongariro River 50, 51, 72, 73, 76, 78, 95-105

In the "correlation and samples taken column", Marker tephra units 1-7 from Chapter 3, and lahar episodes R01-R15 and T1-T4 from Chapter 5 are indicated as well as correlations to other, previously defined tephra formations. Samples taken and referred to in this study are also indicated, as are correlations between the samples.

A 27 Section 3 Upper Waikato Stream T20 465095 True left bank of southern-most tributary of the Upper Waikato Stream, 50m east of State Highway 1, approx 2 km north of Tukino Skifield access road.

Unit Depth Cum. Correlation Description (mm) depth (m) and samples taken

300-500 Taupo lgnimbrite White fine pumice ash with intermixed pumice lapilli and blocks. Occasional charcoal fragments present. Undulating basal contact. 280 0.78 Papakai Formation Unconformable upper contact. Greasy, fine ash, dark greyish brown (2.5Y 4/2), with vertical crack development and evidence of root channels. 90 Reworked red-stained, olive brown (2.5Y 4/4), and grey finelithic and pumice lapilli 140 1.01 Mangamate Tephra (MT), Normally graded coarse ash and fine lapilli. Grey and olive brown Poutu Lapilli poorly vesicular pumice. 160-190 Reworked coarse ash and fine pumice and lithic lapilli grey and reddish brown, crossbedded in places. 110-140 MT, Wharepu Tephra Coarse grey and olive brown ash and fine lapilli, poorly vesicular pumice. Upper contact erosive. Lower 50 mm reddish brown. 20 Yellow brown fine ash 10 MT, Ohinepango Tephra Dark grey med-coarse ash 90 . Fine pumice lapilli, olive brown, dark grey and yellowish red (5YR 4/8) 130 MT, Waihohonu Lapilli Banded layers of med-coarse lithic and poorly vesicular ash, bands dominated by grey, yellowish red, and olive brown. 180 1.77 MT, Oturere Lapilli Dark grey, olive brown and yellowish brown (10YR5/6) ungraded med pumice and lithic lapilli. 120 Dark yellowish brown fine ash with scattered yellowish brown pumice. 50 Pahoka Tephra Greyish brown coarse ash and fine lapilli, some pale brown banding evident within some of the poorly vesicular pumice. 80 Brown and greyish brown med-fine ash with scattered lapilli interbedded. 80 2.10 Bullot Formation (BT), Coarse-fine pale yellow and yellowish brown pumice lapilli and grey Pourahu Member lithic lapilli. 20 Grey coarse ash with occasional lapilli interbedded. 20 Yellowish brown fine pumice angular lapilli and grey lithic coarse ash. 30 Olive grey med-coarse ash and occasional fine pumice lapilli. 10 Yellow brown fine ash 30 grey medium lithic ash 40 Brown fine-med pumice lapilli and grey lithic lapilli. 20 Yellowish brown fine angular pumice lapilli and coarse ash. 20 Greyish brown coarse pumice ash. 50 Yellowish red fine-med pumice lapilli and occasional grey lithic coarse ash. 40 Dark purplish black lithic coarse ash and fine lapilli. 40 Brown pumice med-fine pumice lapilli reversely graded with coarse ash at the base. 10 2.43 BT, M1 Greyish brown med-fine ash 10 Yellowish brown coarse ash and fine pumice lapilli. 10 Fine-med yellowish brown ash with occasional finepumice lapilli. 20 Purplish black coarse lithic ash and fine lapilli. 40 Greyish brown coarse ash and finepumice and lithic lapilli. 20 Purplish grey coarse ash and finelithic lapilli. 10 Strong brown soft pumice lapilli and coarse ash. 20 Purplish black coarse ash and finelithic lapilli. 50 2.61 BT, L 17? Strong brown med-fine pumicelapilli with occasional grey lithic fine lapilli intermixed. 20 Strong brown pumice coarse ash and fine lapilli. 60 Dark greyish brown coarse ash and occasional lithic lapilli. 100 2.79 BT, L 16? Yellowish red (5YR 5/6 ) pumice fine-med lapilli and few grey finelithic lapilli. 10 Grey med-coarse lithic ash. 10 Greyish brown fine ash. 10 Purplish grey med lithic ash. 10 Strong brown fine pumice lapilli and coarse ash. 10 Purplish black med-fine lithic lapilli. 50 BT, M3? Yellowish brown coarse ash and fine pumice lapilli, reversely graded. 20 Purplish black med-coarse ash. 30 Greyish brown coarse-med ash and fine lithic lapilli. 80 3.02 Yellowish red pumice fine-med lapilli with few grey finelithic lapilli. 20 Grey coarse lithic ash. 50 BT, L14? Normally graded greyish brown coarse pumice ash and fine lapilli. 50 i h ash finelapilli lowerGrey s10brown mm fine coarse-med lapilli dominant. with occasional intermixed,

A 28 40 Dark purplish black coarse ash and fine lapilli. 20 Yellowish brown coarse pumice ash and fine lapilli. 20 Purplish black coarse ash and fine lithic lapilli. 20 Greyish brown pumice coarse ash. 10 Purplish black coarse lithic ash. 20 Brown fine pumice lapilli. 10 Grey med lithic ash. 20 Yellowish brown med ash with occasional pumice lapilli. 10 Grey med lithic ash. 30 Greyish brown med-coarse ash mixed with many fine pumice lapilli. 20 Brownish yellow pumice fine lapilli and coarse ash, with few grey lithic lapilli. 80 3.44 Rerewhakaaitu Tephra Greyish brown coarse pumice and lithic ash and fine lapilli. White fine ash interbedded within the base 5-1 0 mm thick. 10 grey coarse lithic ash. 10 Greyish brown med ash and fine pumice lapilli. 10 Grey coarse lithic ash. 30 Yellowish brown pumice fine-med lapilli and few grey lithic fine lapilli. 50 Grey med-coarse ash variable lower boundary. 100 3.65 BT, L7? Yellowish brown coarse pumice ash to med lapilli, shower bedded. 20 Grey med lithic ash. 850 4.50 Hinuera Formation, Bedded silts sands and pebbles, sub-rounded and poorly sorted as a Sample 93.3 whole. Dominantly greyish brown in colour with clasts of pumice and grey lithic andesite. Thin layers of reworked white and pink fine rhyolitic ash interbedded in parts of unit. 1000 5.50 Dark grey andesite lithic med-coarse sands laminated and cross bedded. Layers and lenses of cobbles and pebbles interbedded. Lower contact is an angular unconformity. 1200 R10 Greasy silty matrix diamicton. Matrix supported with pebble to boulder sized clasts. Clasts are heterolithologic, often soft and multi-coloured and appear hydrothermally altered. Lower part of unit contains large (1 m) diameter boulders. Matrix colour yellow to yellowish brown. 250 R10 Similar to unit above, except matrix is pale yellow and there are no boulder sized clasts. 3000+ 9.95 Bedded sands silts and gravels, in lenses and layers. Med-coarse sands and fine pebbles sub-rounded and sub angular. Base into stream.

Section 4 Upper Waikato Stream, T20 466095 True left bank 50m downstream from Section 3. Cover-beds are the same as section 3 and this description is taken from the base of the distinctive R1 diamicton. 0 Unit Depth Cum. Correlation and Description (mm) depth (m) samples taken

100 c.7.0 Bedded silt and sand, with yellow and greyish brown pebbly lenses. 50 Gravels with intermixed sand. 50 Dark red and yellowish brown coarse pumice ash and finelapilli. Normally graded with grey lithic coarse ash interbedded toward the base. 2000-2500 Bedded sands silts and gravels, rounded heterolithologic clasts. Cross-stratified in places, colours range from dark brown to greyish brown sands. 50 9.65 (=93.24) Coarse pumice ash and fine lapilli, ungraded, soft, yellow and pale yellow. 200 Fine grey sand with occasional interbedded pumice pebbles. 150 (=93.25) Reverse graded brown and olive brown fine-med pumice lapilli. 10 Brownish grey fine ash. 400 Grey and pale olive brown coarse-med ash alternating layers of colour dominated by the grey lithic or olive brown pumice ash. Shower bedded. 100 (=93.26, Marker Unit 1) Bedded coarse ash and fine lapilli, strong shower-bedding with grey and olive brown colours on a 30 mm scale. 10-150 10.66 (=93.27, Marker Unit 1) Yellow fine pumice lapilli. 2000 Bedded silts, gravels and sands, brown and greyish brown. Beds and lenses of sub-rounded pebbles and occasional cobbles. Multiple coloured hetero-lithologic pebbles, highly weathered. Laminated and cross bedded. Basal 200 mm gravelly. 500 Purplish grey and purple silty matrix diamicton. Sparse multi-coloured clasts within matrix and mostly of pebble size. Few large boulders near base of unit. 200 (=93.28 and 93.30) Strong brown and yellow, soft, pumice fine lapilli, with few intermixed grey lithics. 30 Purplish grey fine ash.

A 29 50 (=93.29 and 93.31 ) Coarse-med pumice ash, strong brown top and base mostly grey lithic ash. 2000+ 15.44 Silt sand and gravel beds, 50 mm scale beds mostly greyish brown and brown in colour. Lenses of pebbles and occasional small cobbles. Cross-bedded and planar bedded. Sharp contactonto lithified diamicton unit.

Section 5 Upper Waikato Stream, T20 466096 True left bank 100 m downstream from Section 4. Description taken from lowest andesitic tephra described at Section 4.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

40 14.00 93.30 Yellow and yellowish brown pumice fine-med lapilli. Reversely graded. 10 Grey fine ash. 300 93.31 Shower bedded fine-med pumice lapilli, yellowish brown and grey. 20 mm lithic rich layer of fine lapilli. 1000 15.31 Bedded silts sands and gravels, in layers and lenses. Grey and greyish brown in colour with multi-coloured pumice and lithic pebbles. 1200 R1 1 Sharp upper contact. Sandy-matrix diamicton, matrix of fine-coarse sand supporting clasts of pebble to boulder in size. Yellow pumice clasts in finer size range and grey lava in larger size range. Highly lithified unit. Grey and greyish brown matrix colour. 600 Sharp contacts with upper and lower units. Sandy matrix diamicton with maximum clast size of cobbles and pebbles. Grey and greyish brown matrix. 400 Sandy-matrix diamicton, fine grained, clasts of mostly pebble in size. Sharp contact with units above and below. 1500 19.01 Sandy-matrix diamicton, with dominantly cobble sized clasts supported within matrix. Crude horizontal fabric preserved in the base of this bed. Grey and greyish brown in colour. base obscured by talus.

Section 7 Upper Waikato Stream, T20 469100 True right bank of southern tributary, 200 m upstream of junction between the two tributaries. Description taken from above Kawakawa Tephra.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

20 c. 9.0 Yellow brown coarse ash and fine pumice lapilli. 30 93.32, Rerewhakaaitu Greyish brown and yellow brown fine lapilli and coarse ash with Tephra pocketing interbedded white fine rhyolitic ash. 10 Yellow brown fine pumice lapilli. 20-200 Pinching and swelling diamicton, coarse-fine sand matrix supporting mostly pebble clasts with occasional boulders. Unconformable lower contact, undulating. 20 Yellow brown fine pumice lapilli. 10 Grey medium lithic ash. 10 Yellow brown fine greasy ash. 50 Strong brown fine-med pumice lapilli with few grey lithic lapilli. 20 Grey and brownish grey medium ash. 40 9.38 93.33, Kawakawa Tephra Fine yellow pumice ash 20 mm at the top of the unit, underlain by 10 mm pink fine ash and 20 mm further pale yellow fineash. 50 Grey and brownish grey bedded med sand and pebbles. 250 Yellow brown and brownish grey loose sands containing yellow pumice pebbles, unbedded. 400 R10 Pink and grey, greasy, silt-matrix diamicton, multi-coloured pumice and weathered lithic clasts supported in the matrix. Cobble and boulder clasts also common. 500 Brown and dark yellow brown greasy silt-matrix diamicton, containing multi-coloured pumice pebble clasts. 700 As above, but separated by a sharp even contact from the above unit. 350 11.73 93.34 (=93.31) Shower bedded strong brown and yellow brown fine-med pumice lapilli. 10 mm zone of grey lithic lapilli 100 mm above base and reversely graded base. 1200 Grey and greyish brown silt, sand and gravel beds, 2-3 cm scale planar bedding and cross-bedded. 200 R1 1 Greyish brown and grey sandy matrix diamicton, coarse-fine sand matrix supporting pebble clasts, grey red and yellow pumice and lithic clasts. 1200 Normally graded grey and greyish brown sandy matrix diamicton, matrix supporting pebble clasts at the top and cobble clasts at the

A 30 base. Clast rich unit with most of clasts grey and lithic, around 10% of clasts yellow pumice pebbles. Strongly lithified. 1500 Three to four layers of finer grained diamicton units. Sandy matrix with clasts of yellow pumice and dominantly grey lithic pebbles. Strings of cobbles seen and a horizontal fabric preserved. Strongly lithified. 2000 17.83 Sandy-matrix diamicton, no yellow pumice clasts entirely grey lithic pebbles. Strings of cobbles and occasionally huge boulders. Several units pinching in and out with sharp contacts between. Firmly lithified. 100 93.36 Dark red and yellow pumice med-fine lapilli, very firm and cemented. 500 Bedded grown and greyish brown med-fine sand with interbedded yellow pumice in layers and lenses. 150 93.37 Reddish brown and red coarsepumice ash firmlycemented, and shower-bedded. 200 Yellow brown and brown fine ash with occasional yellow pumice lapilli intermixed. 200 93.38 Yellow brown and red stained fine-med pumice lapilli, shower-bedded with few grey lithic fine lapilli. 1000 Brown and dark brown fine ash with paleosol development evident. Greasy fine ash with abundant, finely disseminated charcoal present in the upper 300 mm. Abundant root channels seen as well as vertical cracking and coarse blocky structure. Some intermixed grey and yellow pumice and lithic lapilli near base. 150 20. 13 93.39, Marker Unit 2 Grey and dark grey firmly-cemented fine ash. Extremely hard and breaks up into brittle slabs along shower bedding. Accretionary lapilli within two of the shower beds, 2-5 mm in diameter. 600 Brown and brownish grey fine ash, firm, with little paleosol development. 300 93.40 Yellow and strong brown fine-med pumice lapilli, soft, with occasional fine grey lithic lapilli. 1500 R12 Uniformly fine grained diamicton. Greyish brown sandy and silty matrix with grey lithic pebble sized clasts. Distinctive content of around 20 % yellow pumice fine pebbles. Horizontal fabric observed and unit is firmly lithified. Base of section into stream.

Section 9 Upper Waikato Stream, T20 4681 02 True left bank of north tributary, 40 m upstream of junction of two tributaries. Description taken from above R10.

Unit Depth Cum. Correlation and Description (mm) depth (m) samples taken

1500 c.5.0 Grey and yellow brown bedded fine-coarse sands and fine gravels, unconsolidated, with planar 10-50 mm scale bedding. 600 R10 Yellow brown greasy silt-matrix diamicton. Pebble-cobble clasts supported by the matrix. Multi-coloured pumice and weathered lithic clasts. 1500 Yellow brown and pale yellow silt-matrix diamicton with large grey and red boulder clasts as well as multi-coloured pebble clasts supported by the matrix. 3000 Grey, brownish grey and brown bedded sands, silts and gravels, planar and cross bedded layers, bedding on a 10-100 mm scale. Yellow and yellow brown pumice pebbles common scattered within silts and sands as well as in lenses. 300 10.40 93.29 Yellow brown and yellow fine-med soft pumice lapilli, normally graded with scattered fine grey lithic lapilli. 4000 Bedded silts, sands, and gravels, brown with grey and multi-coloured pumice and lithic pebbles. Planar and cross-bedded on a 50 mm scale. 1500 15.90 R12 Grey and greyish brown sand- and silt-matrix diamicton. Grey lithic pebbly clasts and abundant (>20%) yellow pumice clasts supported by the matrix. Weak planar fabric and firmly lithified. 20 Yellow brown fineash with mixed in fineyellow pumice lapilli. 20 Grey lithic fine lapilli with yellow brown fine ash intermixed. 120 93.41 Strong brown and yellow brown fine pumice lapilli with intermixed grey lithic lapilli. 1000-1500 R12 Firmly lithified diamicton. Grey and greyish brown sandy matrix supporting huge boulder clasts. Abundant large clasts as well as up to 60% of matrix made up of yellow pumice lapilli. 500 93.42 Yellow and yellow brown fine pumice lapilli ungraded and shower bedded. Occasional grey lithic fine iapilli also present. 100 Grey and greyish brown fine ash, firm. 40 Yellow brown and strong brown fine-med pumice lapilli. 150 Grey and greyish brown fine ash with 5% scattered fine yellow pumice lapilli.

A 31 500 18.85 Reddish brown fine ash firm with paleosol development. Abundant brown stained root channels and rare finelydisper sed charcoal. 50 93.43 Yellow and strong brown fine-med pumice lapilli with abundant dark grey lithics and coarse ash sized black free pyroxene crystals. 150 Massive brown fine ash with few scattered fine yellow pumice lapilli, common root channels. 20 Yellow and yellow brown fine pumice lapilli with abundant coarse black free pyroxene crystals. 100 Reddish brown fine ash with common root channels. 40 93.44 Yellow and yellow brown fine pumice lapilli intermixed with grey lithic lapilli, 5-10% free black coarse ash sized pyroxene crystals. 250 Reddish brown fine ash with scattered fineyellow pumice lapilli and black free pyroxene crystals, also common root channels. 350 Grey and greyish brown fineash firm with common yellow pumice and grey lithic lapilli scattered throughout. 350 Reddish brown and brown fine ash, greasy vertical cracking and some blocky structure, common root channels. 50 93.45 Yellow and yellow brown fine pumice lapilli with intermixed grey fine lithic lapilli. 400 R13 Firm diamicton unit, silt and sand matrix supporting mostly grey lithic pebble clasts, also less common pebbles of red and yellow pumice. Erosive lower contact. 50 93.46 Yellow brown and strong brown fine pumice lapilli with intermixed grey lithic lapilli. 300 19.96 93.47, Marker Unit 3 Strong brown stained fine pumice lapilli and coarse ash shower bedded with two distinctive grey lithic dominated zones 10 mm thick 20 mm apart, and 50 mm from the top of the unit. 100 Grey fine ash with scattered fineyellow pumice and grey lithic lapilli. 50 93.48 Yellow brown and strong brown fine pumice lapilli with intermixed grey lithic fine lapilli and black coarse ash sized free pyroxene crystals. 30 Grey fine ash, firm. 50 93.49 Yellow brown and strong brown fine pumice lapilli with intermixed grey lithic fine lapilli and black coarse ash sized free pyroxene crystals. 10 Brown fine ash 200 93.50 Yellow brown and strong brown fine-med pumice lapilli with intermixed grey lithic fine lapilli and black coarse ash sized free pyroxene crystals, concentrated toward the top of the unit. 50 Grey fine ash with scattered 5% yellow pumice lapilli and black pyroxene crystals. 50 93.51 Yellow and yellow brown med-fine pumice lapilli with intermixed grey lithic fine lapilli. 200+ Grey and pinkish grey fineash with scattered 5% yellow pumice lapilli and black pyroxene crystals. 20.70 Base into stream.

Section 10 Upper Waikato Stream, T20 468103 True right bank of stream 50 m downstream of the junction of the two tributaries. Description taken from above the R13 diamicton in section 9.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

50 c. 26.5 Yellow and strong brown fine pumice lapilli with grey lithic coarse ash and black free pyroxene crystals intermixed. 50 Brown and reddish brown fine ash with root channels. 800 R13 Sandy-matrix diamicton. Upper 0-400 mm reworked and partially bedded lower part massive with grey lithic pebble and cobble clasts supported in matrix of grey coarse-med sand, uncemented. 100 Olive brown and grey fine ash, firm. 200 27.65 Marker Unit 3 Strong brown fine pumice iapilli and coarse ash with distinctive 10 mm thick grey lithic dominated layers 20 mm apart 40 mm from the top of the unit. 50 Grey greasy fine ash. 100 (=93.48) Strong brown medium pumice lapilli with intermixed grey lithic lapilli. 50 Grey greasy fine ash. 200 (=93.50) Yellow and strong brown fine pumice lapilli with mixed grey finelithic lapilli. 50 Greyish brown greasy fine ash. 30 (=93.51 ) Yellow and pale brown fine pumice lapilli with grey lithic fine lapilli and black free pyroxene crystals intermixed. 70 Grey fine ash with scattered fineyellow pumice lapilli intermixed. 200 Brown fine ash firm with root channels. 100 93.52, Marker Unit 4 Pale brown and strong brown med-coarse pumice lapilli with grey lithic fine lapilli and black free pyroxene crystals intermixed.

A 32 200 Bedded silt and sand with intermixed yellow pumice fine lapilli. 60 93.53 Strong brown fine pumice lapilli and coarse ash. 10 Grey fine lithic ash. 30 93.54 Strong brown and reddish brown coarse-med pumice lapilli. 150 Alternating 30 mm layers of olive brown and yellow fine pumice lapilli. 150 Grey lithic coarse ash and strong brown fine lithic lapilli. 120 Pale brown and strong brown alternating 30 mm layers of fine pumice lapilli. 200 Many thin (10 mm) layers of brown , grey, yellow, and pale brown fine pumice lapilli and coarse ash. 1200 31.52 R14 Sandy-matrix diamicton. Matrix supporting grey clasts of pebble to small boulder in size. Brown to greyish brown clasts and matrix. Ungraded and with no bedding. Sharp planar lower contact. 100 Dark grey and strong brown coarse-med pumice lapilli in a matrix of grey coarse ash. 50 Grey coarse lithic ash and fine lapilli. 100 Brown fine pumice lapilli and coarse ash. 90 Pale brown and grey coarse ash and fine pumice lapilli. 50 93.58 Strong brown and pale brown fine pumice lapilli with intermixed grey lithic fine lapilli. 200 Sandy matrix diamicton, pebble clasts supported in greyish brown med-coarse sand. 20 Pale brown and yellow fine pumice lapilli. 1500+ 33.63 R15 Grey sandy-matrix diamicton, pebble and cobbleclasts supported in med-coarse sand matrix, unbedded and uncemented. Base into stream.

Section 11 Tangatu Stream, T20 4701 10 True left bank of stream, 500m downstream of state highway 1.

Unit Depth Cum. Correlation and Description (mm) depth (m) samples taken

400 Makahikatoa Sand Brown and greyish brown med-fine sand, present soil development. 450 Taupe lgnimbrite White fine ash matrix supporting white pumice block and lapilli clasts. Charcoal present near base. 300 1.15 Mangatawai Tephra Brown and greyish brown fine greasy ash with several black and purplish grey fine-coarse ash beds. Yellowish brown beech leaves preserved in some of the ash layers. 100 Papakai Formation Brown and yellow brown fine greasy ash with strong paleosol development. 30 1.28 Waimahia Tephra Cream-cakes of fine white ash. 300 Papakai Formation Yellow brown fine greasy ash. 150 1.73 Hinemaiaia Tephra within Coarse white pumice ash scattered throughout brown fine ash. Papakai Formation Unconformable lower contact. 1200-1500 Bedded sands, silts and gravels. Greyish brown and brown, planar bedded and cross bedded, with layers and lenses of grey lithic and yellow pumice pebbles. 10 cm scalebedding. 1500 Brown silt-matrix diamicton. Large boulder clasts supported within matrix as well as grey pebbles and cobbles. 500+ R10 Yellow brown sandy and silty matrix diamicton. Fine grained with dominantly pebble clasts. Firm matrix supporting multi-coloured pumice and weathered lithic clasts. 5.23 Base into stream.

Section 14 Upper Waikato Stream, T20 4771 12 True left bank of stream, 300 m upstream of junction with Tangatu Stream. Description taken from Bullot Formation member containing Rerewhakaaitu Tephra.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

50 c. 7.0 93.59, Rerewhakaaitu Greyish brown coarse ash and fine lithic lapilli containing cream-cakes Tephra of up to 20 mm thick of white fine rhyolitic ash. 30 Strong brown and yellow brown fine-med pumice lapilli. 10 Grey coarse ash. 0-50 Eroded yellow brown and strong brown fine, soft pumice lapilli. 400-500 Grey and greyish brown bedded sands and gravels in planar layers and lenses. Variable thickness and unconformable contacts. 400 93.60 Yellow brown and strong brown fine-med pumice lapilli with few grey fine lithic lapilli interbedded. 2000 9.99 R1 O/R1 1 Brown silt and sand matrix diamicton. Matrix supports clasts of pebble to boulder in size. Multi-coloured lithic and pumice clasts abundant.

A 33 Massive and unbedded, boulders concentrated toward the base. 10 Yellow medium pumice lapilli in a matrix of brown fine ash. 20 Dark greyish brown fine ash. 20 Yellow coarse pumice ash and fine pumice lapilli. 30 Grey fine ash. 20 Yellow medium pumice lapilli. 30 Grey fine ash. 50 Dark greyish brown fine ash with scattered fineyello w pumice lapilli. 200 93.61 (=93.37) Olive grey coarse ash, shower-bedded and ungraded. 50 Brown fine ash with scattered yellow fine-pumice lapilli and coarse ash. 100 93.62 (=93.38) Yellow brown fine and soft pumice lapilli. 1000 11.49 Brown and greyish brown fine ash, greasy and with common root channels. Finely disseminated charcoal present in the upper 200 mm. Weak blocky structure developed near top. Lower 200 mm contains scattered yellow and yellow brown finepumice lapilli (5%). 150-200 Marker unit 2 Pale grey and light brownish grey hardened fine ash tuftunit. Shower bedded fine and med ash. Breaks into brittle slabs along beds. 400 93.63 (=93.40) Yellow and yellow brown fine-med pumice lapilli with intennixed grey fine lithic lapilli. 150 Brown fine ash with scattered yellow brown fine pumice lapilli. 200 93.64 (=93.40) Yellow brown med-fine pumice lapilli and grey lithic finelapilli. 1400 13.84 R12 Massive diamicton. Silt and sand-matrix supporting pebble to boulder clasts. Abundant grey lithic and yellow pumice clasts. Finnly lithified. 50 93.65 (=93.42) Yellow and brown fine pumice lapilli and coarse ash. 350 Greyish brown fine ash with scattered coarse yellow pumice ash (10%) 50 Yellow brown finepumic e lapilli and coarse ash. 20 Greyish brown fine ash. 40 93.66 Yellow and yellow brown med-fine pumice lapilli. 200 Greyish brown fine ash with abundant root channels and few scattered yellow fine pumice lapilli. 100 93.67 Yellow and strong brown med pumice lapilli, with grey fine lithic lapilli. 50 Reddish brown fine ash with root channels. 50 93.68 Yellow fine pumice lapilli with grey fine lithic lapilli and black coarse ash sized pyroxene crystals. 200 Dark red fine ash with scattered fine yellow pumice lapilli (5%), and root channels. 50 15.00 93.69 Yellow pumice lapilli with abundant black coarse ash grade pyroxene crystals. 400 Dark red fineash with scattered fine yellow soft pumice lapilli, some root channels 150 Yellow and grey fine pumice lapilli mixed with finebrown ash. 100 Dark red fine greasy ash with scattered fineyellow soft pumice lapilli, some root channels. 1000-1200 16.85 Bedded silts and fine sands, laminated and cross-bedded, lenses and layers rich in pebbles. Dusky red and greyish brown. 20 Grey and Yellow brown coarse ash. 300 Pink fine ash, finn, and contains root channels. 300 R13 Firm silt matrix diamicton, grey red and yellow pebble to cobble-sized clasts supported by the matrix. 150 Greyish brown fine ash with intennixed yellowbrown fine pumice lapilli. 200 17.82 93.70 (=93.47), Red coarse pumice ash and fine lapilli with grey lithic lapilli and black Marker unit 3 coarse ash grade pyroxene crystals intennixed. 3 distinctive grey 10 mm layers separated by 20 mm near the top of the unit. 70 Grey fine ash with scattered fineyellow pumice lapilli. 200 93.71 , (=93.48) Strong brown and yellow brown med-fine, soft, pumice lapilli with grey lithic lapilli and black coarse ash grade pyroxene crystals intennixed. 50 Greyish brown and grey fineash with scattered fineyellow pumice lapilli. 250 93.72, (=93.49) Yellow brown and pink fine-med pumice lapilli with grey lithic lapilli and black coarse ash grade pyroxene crystals intennixed. 70 Grey fine ash. 50 93.73, (=93.51 ) Yellow and yellow brown fine-med pumice lapilli. 100 Grey and greyish brown fineash with 5% yellow fine pumice lapilli intermixed. 50 18.66 93.74, Marker Unit 4 Yellow brown pumice lapilli with intennixed grey fine lithic lapilli. 10 grey coarse ash. 50 93.75 Yellow and greyish brown and black pumice and lithic fine lapilli, lithic rich unit. 200 Brown and greyish brown fine ash with scattered fineyellow pumice lapilli. 20 Yellow pumice coarse ash and fine lapilli with intennixed grey fine lithic lapilli. 200 Greyish brown fine ash with scattered fine yellow pumice lapilli.

A 34 20 Brown and yellowish brown fine pumice lapilli. 30 Grey med ash and greyish brown and black coarse lithic ash. 100 19.29 93.76 Brown and strong brown fine-med pumice lapilli, firm lapilli. 40 Greyish brown med-coarse ash and grey lithic fine lapilli. 50 93.77 Yellow and yellow brown fine pumice lapilli with intermixed fine grey lithic lapilli. 100 Brown and yellow brown fineash with scattered yellow brown fine pumice lapilli. 200 Brownish grey med-coarse ash with scattered grey lithic and yellow pumice fine lapilli. 30 Brown fine ash with scattered yellow brown fine pumice lapilli 20 Grey coarse ash. 30 Brown coarse ash. 30 Yellow brown fine pumice lapilli wi th intermixed grey coarse lithic ash. 30 Brown coarse ash. 10 Dark grey lithic med ash. 40 19.87 93.78 Yellow and yellow brown fine pumice lapilli with 50% intermixed grey lithic fine lapilli. 10 Grey fine lithic ash. 30 Yellow and yellow brown fine pumice lapilli with intermixed grey coarse lithic ash. 40 Grey med ash with scattered yellow fine pumice lapilli. 20 Brownish grey fine ash with scattered yellow fine pumice lapilli. 40 Grey med ash with scattered yellow fine pumice lapilli. 250 Brownish grey med ash with scattered yellow fine pumice lapilli. 20 Yellow brown fine pumice lapilli with intermixed grey lithic coarse ash. 250 Greyish brown and grey fine-coarse ash with scattered yellow fine pumice and grey lithic lapilli. 100 20.63 93.79 Yellow brown strong brown, greyish brown and black med-coarse pumice lapilli. 200 93.80 Strong brown and yellow brown fine-med pumice lapilli. 10 Brown firm, fine ash. 100 Brown and greyish brown med-coarse ash and fine pumice lapilli. 50 93.81 Brown and yellow brown fine-med pumice lapilli with grey fine lithic lapilli. 1500 R14 Sandy matrix diamicton, grey and brownish grey med-coarse sand supporting grey and red lithic pebbles-boulders. Poorly sorted, unconsolidated and unbedded. Some faint planar fabric. 600 Bedded sands and pebbles, 30 mm scale planar and cross bedding. Greyish brown with lenses of yellow and strong brown pumice pebbles. 50 23.14 93.82 Black stained yellow fine-med pumice lapilli. 100 Grey and brownish grey med-fine ash with scattered fineyello w pumice lapilli. 40 93.83 Black stained, yellow brown fine-med pumice lapilli. 400 93.84 Greyish brown, yellow and grey fine-med shower-bedded pumice lapilli. 50 Grey med coarse-fine ash. 20 Brownish grey fine ash. 50 93.85, Marker Unit 5 Normally graded yellow and grey pumice and lithic fine lapilli and coarse ash. 500 Grey bedded sands and gravels in lenses and layers with yellow pumice pebbles. 1000 R15 Grey and greyish brown diamicton. Coarse-fine sand matrix supporting pebble-cobble cfasts and some small boulders. Planar fabric preserved with cobblestrings. 200 25.50 93.86 Brown, brownish grey and yellow mad-coarse pumice lapilli, shower bedded, lower 100 mm has intermixed grey med ash or sand. 5000 R15 Several 1 m thick diamicton units interbedded with bedded silts sands and gravels in layers and cross beds. Diamictons have matrices of coarse-med sand supporting pebble-cobble cfasts with occasional boulder. Unconsolidated and overall planar fabrlc preserved with cobble strings. Dominantly grey and brownish grey lithic with occasional red lithic clasts. 2500 Coarse grained diamicton. Coarse-mad grey sand matrix supporting large grey angular boulder and cobble cfasts. Weak planar fabric. 5000 Grey and greyish brown planar diamictons and interbedded finely bedded sands and gravels. Sandy matrix supporting lithic pebble and cobble clasts. 5000 Bedded sands and gravels, grey and brownish grey, planar and cross bedded in layers and lenses. Upper half dominantly gravels and lower half dominantly sands. 1000 Pale brownish grey firm, fine sand, massive unbedded with 5 % scattered fine yellow pumice lapilli. 500-1000 Variable thickness dark grey diamicton. Sandy matrix supporting

A 35 pebble-boulder lithic clasts. 1000 Grey and olive grey well sorted planar bedded med sand. 50 Grey coarse-med ash. 50 46.10 93.87 White and strong brown stained fine pumice lapilli. 100 93.88 Pinkish white and grey fine pumice lapilli. 300 Fibrous dark brown and black firm lignite. Finely bedded 10 mm scale, planar bedding. Sampled at 50 mm intervals. 20 93.90 White and pale grey med ash. 300 Fibrous pale brown and dark brown lignite planar laminated. 40 93.89, Marker Unit 6 Grey and pinkish grey coarse ash and fine pumice and lithic lapilli. 20 Brown lignite. 100 Finely laminated brown lignite and grey fine-med sand. 150 Dark brown and pale brown laminated lignite. 100 93.91 , Marker Unit 7 Grey and pinkish grey med pumice ash. 400 Bedded fine-coarse pale grey sands. 1000+ R15 Coarse sand matrix diamicton, large boulder-pebble clasts supported within the matrix. Brownish grey unbedded, massive and very firmly consolidated. 48.63 Base into stream.

Section 17 Wharepu Stream, T20 463122 True left bank of stream, 50 m upstream of western power pylons, approximately 500 m downstream of State Highway 1.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

800 Ngauruhoe Formation Brown and greyish brown fine-med ash, friable, with present soil development. 100-250 Taupo lgnimbrite White fine-med pumice ash supporting ciasts of white pumice lapilli and blocks. Charcoal fragments dispersed throughout. 300 Mangatawai Tephra Dark brown and brown fine greasy ash with paleosol development and paleo-root channels. The lower 100 mm contains several 10-20 mm thick fine-med purplish black and black ash beds. Yellow brown beech leaves are preserved within some of these layers. 100 Papakai Formation (Pp) Yellowish brown fine greasy ash, friable, with paleosol development. 20 1.47 Wa imahia Tephra within Cream-cakes of greyish white fineash interbedded within yellow brown Pp greasy fine ash. 150 Pp Brown greasy fine ash. 100 1.72 Hinemaiaia Tephra within Scattered med-coarse white pumiceous ash within brown greasy fine Pp ash. 200 Pp Brown fine greasy ash. 40 1.96 Motutere Tephra within Pp Pale brown med-fine ash in cream-cakes within brown greasy fineash. 50 Pp Brown fine greasy ash, strong brown staining at basal contact. 30 Grey coarse ash. 50 2.09 Mangamate Tephra (MT), Strong brown and grey lithic and poorly vesicular pumice lapilii with Poutu Lapilli grey coarse ash base. 500 MT. Wharepu Tephra Shower-bedded fine-med poorly vesicular pumice and lithic lapilli. Grey, strong brown and greyish brown . Strong brown and reddish brown basal 30 mm of fine lapilli, distinctive. 20 Yellow brown fine firm ash. 10 Dark grey coarse ash. 260 MT. Waihohonu Lapilli Grey brownish grey and strong brown fine pumice and lithic lapilli. Strong shower bedding and alternating 10-50 mm layers dominated by the different colours. 400-500 MT. Oturere Lapilli Coarse-med grey, brownish grey and strong brown pumice and lithic lapilli, shower-bedded but ungraded. 150 Yellow brown fine ash with yellow pumice fine lapilli scattered through a zone in the middle of the unit. 50 3.58 Pahoka Tephra Distinctive platy olive grey fine pumice lapilli. 100 Dark yellowish brown fine ash with scattered fine yellow pumice lapilli. 50 3.73 Bullot Formation (BT), Pale brown coarse pumice lapilli. Pourahu Tephra 50 Normally graded coarse ash-med lapilii, yellow pumice and grey lithic lapilli intermixed. 50 Brownish grey coarse ash and fine lapilli. 10 Yellow brown fineash and scattered fine pumice lapilli. 10 Grey lithic coarse ash. 50 Pale brown pumice and grey lithic med lapilli 30 Grey med lithic lapilli. 30 3.96 BT, M1? 10 mm olive grey med ash over 10 mm yellow fine pumice lapilii, over 10 mm grey coarse ash. 40 Brownish grey fine-med iapilli. 30 Yellowish brown fineash with scattered coarse pumice and lithic yellow and grey ash.

A 36 50 Strong brown and grey pumice and lithic med-coarse lapilli. 20 Yellowish brown fine pumice lapilli. 30 Grey lithic fine lapilli. 10 Yellow brown coarse pumice ash. 10 Grey fine lithic lapilli. 30 Strong brown and grey med pumiceand lithic lapilli. 20 Brownish grey coarse pumice ash and fine lapilli. 100 4.30 BT, L 17? Strong brown and grey pumice and lithic med lapilli shower-bedded and ungraded. 30 Brownish grey pumice coarse ash and fine lapilli. 50 Brownish grey and grey fine pumice and lithic lapilli. 30 Dark grey coarse ash. 30 Brownish grey and grey coarse ash and finelapilli. 120 4.56 BT, L16? Strong brown coarse pumice lapilli with intermixed grey med lithic lapilli. 30 Brownish grey and grey coarse pumice and lithic ash. 50 Strong brown fine-med pumice lapilli and grey lithic fine lapilli. 150 4.79 BT, L 15? Strong brown and grey med pumice and lithic lapilli shower-bedded but ungraded. 0-1 0 Waiohau Tephra Pocketing fine, white ash. 30 Brownish grey fine pumice lapilli. 20 Grey coarse ash. 50 Brownish grey, strong brown and grey fine pumice and lithic lapilli, normally graded. 20 Grey coarse lithic ash. 150 BT, L14 Strong brown med pumice lapilli, shower-bedded but ungraded. 50 Brownish grey coarse-med pumice ash. 10 Grey coarse ash. 4000+ Grey andesitic lava flow. 9.10 93.106 Base into stream.

Section 18 Wharepu Stream, T20 465123 True left bank of stream 300m downstream of Section 17. Description taken from below L 14? in Section 17.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

1000 c. 5.0 Reworked ash and lapilli, in cm scale beds and layers and lenses. Grey brownish grey and pale olive brown pumice and lithic pebbles. 50 Dark purplish grey coarse ash and fine lapilli. 200 Reworked brownish grey coarse ash and fine lapilli wavy bedding. 40 Brownish grey fine lapilli. 10 Grey fine lapilli. 10 Brownish grey fine ash. 30 Reworked Brownish grey coarse ash and lapilli, wavy, finebedding. 30 Strong brown and grey fine pumice and lithic med lapilli. 150 Reworked brownish grey coarse ash and finelapilli mm scale beds and laminations. 20 Strong brown med pumice lapilli and grey lithic finelapilli. 40 Reworked brownish grey ash and lapilli. 200 5.78 Reworked dark purplish grey coarse ash and fine lapilli, wavy and planar bedding. 40 Pale olive brown med-fine pumice lapilli. 400 Brownish grey reworked med-coarse ash and fine pumice lapilli mm scale wavy and planar beds. 250 Dark purplish grey reworked coarse lithic ash in planar cm scalebeds. 50 6.52 Dark purplish grey fine lithic lapilli and olive brown pumice finelapilli. 20 Olive brown coarse pumice ash and grey lithic coarse ash. 30 Grey coarse ash. 150 Strong brown med-coarse pumice lapilli with intermixed grey lithic fine lapilli. 250 Reworked brownish grey med-coarse pumice and lithic ash, in planar and wavy bedded layers. 50 7.02 Pale olive brown med pumice lapilli and grey lithic fine lapilli. 50 Grey and olive grey coarse-med ash. 20 Grey lithic med ash. 100 Shower-bedded olive brown and brownish grey coarse ash and fine pumice lapilli. 30 Brownish grey med-fine pumice lapilli. 10 Grey coarse lithic ash. 30 Strong brown fine pumice lapilli and grey finelithic lapilli. 10 Brownish grey fine ash. 50 7.32 Brownish grey coarse pumice ash and fine lapilli. 1300 Reworked grey and pale olive brown coarse ash and fine pumice

A 37 lapilli, planar and cross bedded on a mm scale. 150 Shower-bedded pale olive brown fine-med pumice lapilli. 1200 Grey sandy matrix diamicton. Brownish grey med-coarse sand supporting grey pebble to cobble clasts. Pinches and swells in thickness and has bouldery clasts concentrated in places. 1000+ R10 Silty matrix clast rich diamicton. Yellowish brown and pale brown firm silt matrix. Clasts mostly pebble-cobble in size and are multi-coloured pumice and lithics. 10.97 Base into stream.

Section 22 Te Piripiri Stream, T20 453129 True right bank of southern tributary of stream, 20 m upstream of fork in southern sub-fork.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

500 Taupo ignimbrite White and pale grey fine-coarse pumice ash with pumice lapilli and blocks. Charcoal fragments near base. 500 Mangatawai Tephra Dark brownish grey and brown fine ash with paleosol development, lower half of unit contains several 10-30 mm thick grey and purplish grey ash beds. Abundant yellow brown beech leaves are preserved within some of the layers. 50 1.05 Papakai Formation (Pp) Yellow brown and brown fine greasy ash with paleosol development. 20 Strong brown and yellow fine pumice lapilli and coarse ash. 50 Yellow brown fineash. 20 1.14 Waimahia Tephra within Cream-cakes of white, fine rhyolitic ash within yellow brown and brown Pp fine ash. 30 Pp Yellow brown fine ash. 100 1.27 Hinemaiaia Tephra within White med-coarse pumice ash speckled within fine brown greasy ash. Pp 400 Pp Yellow brown and brown fine ash with intermixed grey lithic and poorly vesicular pumice fine lapilli near the base. 50 1.72 Mangamate Tephra (MT), Strong brown and grey pumice and lithic fine-med lapilli. Poutu Lapilli 230 MT, Wharepu Tephra Grey and brownish grey fine pumice and lithic lapilli and shower bedded coarse ash. Distinctive strong brown basal 30 mm. 200 MT, Waihohonu Lapilli Alternating coloured zones of grey and strong brown dominated coarse pumice and lithic ash, strong banded shower-bedding. 200 Grey and strong brown finepumice and lithic lapilli and coarse ash, shower-bedded. 300 MT, Oturere Lapilli Brownish grey and grey fine pumice and lithic lapilli and coarse ash. 50 Yellow brown fine ash with scattered fine yellow brown pumice lapilli. 50 Yellow brown fine ash with grey fine ash intermixed. 90 2.84 Pahoka Tephra Olive grey platy blade like pumicefine lapilli and coarse ash. Normally graded onto olive grey and grey fine-med pumice and lithic lapilli. 100 Yellow brown fine pumice ash with scattered yellow brown fine pumice lapilli. 50 2.99 Bullet Formation (BT), Pale yellow and yellow brown coarse-med pumice lapilli. Pourahu Member 50 Pale yellow and grey med-fine pumice and lithic lapilli and coarse ash. 30 Greyish brown fine pumice lapilli and coarse ash with few grey lithic fine lapilli. 30 Olive grey and grey pumice and lithic finelapilli and coarse ash. 10 Pale brown fine ash and scattered fine lithic lapilli. 100 3.21 Brownish grey coarse ash with scattered med grey lithic lapilli. 50 Olive grey and yellow fine-med coarse pumice and lithic ash. 30 Olive brown fine ash and scattered fine pumice lapilli. 30 Grey coarse lithic ash and fine lapilli. 10 Olive brown coarse pumice ash and finelapilli. 20 Grey fine lithic lapilli. 30 Olive brown fine-med pumice lapilli with few intermixed grey lithic fine lapilli. 10 Brownish grey med ash with scattered finepumice lapilli. 50 Pale brown and yellow fine-med pumice lapilli with few grey lithic fine lapilli. 200 3.64 Brownish grey and yellow coarse pumice ash with intermixed grey lithic coarse ash. 30 Yellow brown fine pumice lapilli. 20 Grey fine lithic lapilli and coarse ash. 10 Brown fine ash. 70 Olive grey coarse lithic ash with scattered yellow fine pumice lapilli. 50 Strong brown and yellow fine pumice lapilli with few grey lithic fine lapilli. 100 3.92 Brownish grey coarse ash with scattered grey fine lithic lapilli.

A 38 50 Yellow and pale brown finepumice lapilli with few grey lithic finelapilli. 10 Grey fine lithic lapilli and coarse ash. 100 Yellow and yellow brown fine-med pumice lapilli with few grey lithic fine lapilli. 30 Brownish grey coarse ash with scattered yellow pumice and grey lithic fine lapilli. 30 4.14 93.6, Waiohau Tephra White fine ash, semi-continuous layer with occasional yellow fine pumice lapilli mixed in. 20 Olive brown coarse pumice ash and fine lapilli. 20 Dark brownish grey fine ash with scattered fine yellow pumice lapilli. 50 Dark brownish grey fine pumice lapilli and coarse ash. 30 Grey and olive brown fine pumice and lithic lapilli. 20 Yellowish brown and yellow fine pumice lapilli. 50 Dark grey lithic coarse ash and fine lapilli. 30 Pale yellow med-fine pumice lapilli. 300 4.66 Shower-bedded grey and pale brownish grey coarse pumice and lithic ash. 100 Strong brown and yellow brown fine-med ungraded pumice lapilli. 500 Reworked pale brownish grey, grey and yellow pumice and lithic med coarse ash. Layers of yellow and grey pumice and lithic lapilli also occur. 100 5.36 Pale olive brown med-fine pumice lapilli with scattered grey lithic fine lapilli. 100 Greyish brown coarse lithic ash. 150 Brownish grey coarse ash and fine pumice lapilli. 50 Pale brown fine pumice lapilli with scattered fine grey lithic lapilli. 1000 R10 Sandy matrix diamicton brownish grey matrix supporting clasts from pebble-boulder, clast rich in places. 1000+ Silt and sand-matrix diamicton. Greyish brown and brown matrix supporting heterolithologic, multicoloured clasts from pebble to large boulder. 7.66 Base into stream.

Section 24 Te Piripiri Stream, T20 448137 True left bank of northern tributary stream, 1 km upstream from State Highway 1.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

600 Makahikatoa Sand and Brownish grey fine sand with interbedded 5- 10 mm thick black fine Tufa Trig Formation with med ash beds interbedded throughout sand. At 300 mm depth white Kaharoa Tephra? pocketing fine pumice ash preserved. Present soil development. 200 Taupo lgnimbrite White pumice ash with interbedded pumice lapilli and blocks. 1000 Mangatawai Tephra Dark brown and dark greyish brown greasy fine ash with paleosol development. In basal half of unit preserved several 10-20 mm black and purplish grey fine-med ash beds. Yellow brown beech leaves preserved within the ash. 50 1.85 Papakai Formation (Pp) Greyish brown and yellow brown greasy fineash. 20 Strong brown and pale yellow fine pumice lapilli. 200 Olive grey and greyish brown fine greasy ash. 20 2.09 Waimahia Tephra within Pocketing fine white pumice ash within greyish brown fine ash. Pp 100 Pp Greyish brown fine ash. 100 2.29 Hinemaiaia Tephra within White med-coarse pumice ash speckled within greyish brown fine Pp greasy ash. 20 Pp Strong brown fine pumice lapilli. 1000 Yellow brown greasy fine ash with strong brown and grey finepumice and lithic lapilli intermixed toward the base of the unit. 100 3.41 Mangamate Tephra (MT), Strong brown and reddish brown lithic and poorly vesicular pumice Poutu Lapilli lapilli. 1100 MT, Wharepu Tephra Olive brown and greyish brown coarse pumice and lithic ash and fine lapilli. Shower-bedded and ungraded. Basal 100 mm strong brown fine pumice and lithic lapilli. 300 4.81 MT, Waihohonu Lapilli Strong brown and grey fine-med pumice and lithic lapilli, reversely graded. 10 Brownish grey fine ash and fine pumice lapilli. 50 Dark grey med shower-bedded ash. 60 Yellow brown andlithic greyish brown fine ash and interbedded strong brown fine pumice lapilli. 40 Dark grey med-coarse lithic ash. 400 5.37 MT, Oturere Lapilli Greyish brown, strong brown and grey fine-med pumice and lithic lapilli. 50 Yellow brown fine greasy ash and scattered grey fine lithic lapilli. 30 Pale yellow and grey fine pumice and lithic lapilli and coarse ash.

A 39 40 Pale brown fine ash with scattered fine yellow pumice lapilli. 50 5.54 Pahoka Tephra Olive grey platy fine pumice lapilli. . 20 Greyish brown fineash with interbedded yellow finepumice lapilli. 130 Olive grey platy coarse ash and fine-med pumice lapilli, normally graded. 10 Pale yellow and grey coarse ash. 500 Pale greyish brown silty and sandy matrix diamicton. Pebbles boulders supported within the matrix. Poorly sorted and unbedded, erosive angular unconformity as basal contact. 200 Yellow brown fine ash and interbedded olive brown fine pumice lapilli. 50 6.45 Bullot Formation/ Pourahu Pale yellow med pumice lapilli. Member 100 Olive brown and yellow fine-med pumice lapilli and coarse ash. 1000+ R07 Grey med sandy matrix diamicton. Greyish brown coarse-med sand matrix with clasts of pebble to small boulder. 7.55 Base into stream.

Section 26 Te Piripiri Stream, T20 435129 True left bank of main northern tributary of stream 600 m upstream of where other tributaries join the main stream. Description taken from base of Pahoka Tephra.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

50 c. 3.0 Coarse pale yellow and grey pumice and lithic ash. 30 Olive brown fine pumice lapilli. 100 Yellow brown and pale brown fineash with scattered grey lithic and pale brown pumice lapilli. 50 3.18 Bullot Formation (BT), Pale yellow and pale brown med-coarse pumice lapilli with scattered Pourahu Member grey med lithic iapilli. 100 Pale brown and pale yellow coarse ash and fine pumice lapilli with scattered grey coarse lithic ash. 50 Greyish brown coarse pumice ash with grey lithic lapilli intermixed. 50 Yellow brown and grey pumice and lithic fine lapilli. 20 Grey lithic coarse ash. 20 Greyish brown coarse pumice and lithic ash. 30 Grey and greyish brown fine-med pumice and lithic lapilli. 30 Olive grey coarse-med pumice lapilli. 30 Grey coarse ash and fine lithic lapilli. 10 3.52 BT. M1? Olive brown coarse pumice ash. 10 fine olive grey ash. 10 Grey bedded coarse lithic ash. 100 Reworked pale brown and olive brown fine-medash. 50 Greyish brown and grey coarse pumice and lithic ash and fine lapilli. 30 Pale olive brown fine pumicelapilli and coarse ash with scattered grey lithic coarse ash. 10 Grey med lithic ash and scattered fine lithic lapilli. 30 Olive brown med-coarse pumiceash with scattered fine pumice lapilli. 50 3.79 Olive brown fine pumice lapilli with scattered grey lithic fine lapilli. 30 Yellow brown fineash with scattered yellow brown fine pumice lapilli. 50 Grey and greyish brown coarse pumice and lithic ash and finelapilli. 30 Dark brownish grey fine-med pumice lapilli. 30 Grey coarse ash. 50 Greyish brown and olive brown fine-med pumice lapilli with scattered fine grey lithic lapilli. 300 4.28 Brownish grey silty-matrix diamicton. Pebble clasts supported within the matrix, no bedding. 80 Yellow brown and strong brown med-coarse pumice lapilli with scattered grey lithic fine lapilli. 50 Brownish grey coarse pumice ash and fine lapilli. 100 Yellow and grey fine-med pumice and lithic lapilli. 20 Grey coarse lithic ash with scattered yellow fine pumice lapilli. 150 Grey fine-med ash with scattered yellow and yellow brown fine pumice lapilli. 50 Grey coarse lithic ash and fine lapilli. 150 Grey and greyish brown pumice and lithic, shower-bedded coarse ash and fine lapilli. 20 Yellow and yellow brown fine-med pumice lapilli. 20 5.02 93.8, Rerewhakaaitu Grey coarse ash with scattered yellow pumicelapilli and interbedded Tephra white fine pumice ash 10 mm thick. 100 Reworked purplish grey med ash to fine lithic lapilli. 30 Grey med ash with scattered yellow fine pumice lapilli. 20 Pale brown and yellow pumice coarse ash with grey coarse lithic ash intermixed.

A 40 200 Reworked grey and brownish grey fine-med ash with layers of fine med pumice lapilli within. 50 Yellow brown and olive brown fine-med pumice lapilli with scattered grey fine lithic lapilli. 300 5.72 Reworked black and purplish grey coarse lithic ash and fine lapilli, 10- 20 mm planar bedding. 30 Pale olive brown fine-med pumice lapilli. 10 Grey coarse lithic ash. 10 Strong brown and pale yellow coarse pumice ash. 20 Greyish brown pumice med-coarse ash. 150 Reworked purplish grey and brownish grey lithic and pumice coarse ash and fine lapilli. 70 Strong brown and yellow fine-med pumice lapilli with scattered grey fine lithic lapilli. 150 Brownish grey bedded med-sands, 10-20 mm scale beds. 100 6.26 Pale olive brown and yellow fine pumice lapilli. 400 Brownish grey med sands with layers and lenses of brownish grey and yellow fine pumice lapilli. Planar and cross bedded. 10 Yellow brown silt. 600 Dark purplish grey coarse lithic ash and shower-bedded finelapilli. 200 Pale olive brown and grey fine-med pumice and lithic lapilli. 1000 R09 Brownish grey sandy matrix diamicton. Pebble-boulder ctasts supported within the matrix, grey clasts, unbedded. Angular unconformity as lower contact. 100 Strong brown and reddish brown fine soft pumice lapilli. 1000+ 9.57 R10 Yellow brown silty and sandy matrix diamicton. Pinches and swells over lava flow. Clasts of pebble to huge lava boulders contained within matrix. 10000+ Grey andesite lava flow. 19.57 Base into stream.

Section 30 Te Piripiri Stream, T20 4861 62 True left bank of stream where four-wheel-drive track crosses lower Stream, 1 km upstream from Tongariro River.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

300 Ngauruhoe Formation Brown and brownish grey fine-med ash, present soil development. 1000+ Taupo lgnimbrite White fine-coarse pumice ash with pumice lapilli and blocks. Charcoal fragments near base and pink staining in places. Unit ramps over a hill and thickens in paleo-valley. 300 1.60 Mangatawai Tephra Dark greyish brown fine ash with paleosol development lower half contains several grey and purplish grey fine-med ash, 10-20 mm layers. 100 Papakai Formation (Pp) Yellow brown and brown fineash with few scattered yellow brown fine pumice iapilli 20 1.72 Waimahia Tephra within White fine ash cream-cakes within fine greasy yellow brown ash. Pp 30 Pp Yellow brown fine ash. 50 1.80 Hinemaiaia Tephra within White coarse-med pumice ash scattered within yellow brown fine Pp greasy ash. 50 Pp Yellow brown fine ash. 20 Strong brown fine pumice iapilli. 100 Yellow brown fine ash. 20 1.99 Motutere Tephra within Pp Pale brown and white fine pumice ash cream-cakes within yellow brown fine ash. 50 Pp Yellow brown fine ash with scattered grey pumice lapilli. Strong brown stained base. 200-400 R02 Greyish brown med-coarse and matrix diamicton, angular grey and strong brown pebble-cobble clasts supported within the matrix. Very ciast rich unit with abundant strong brown and pale brown pumice ciasts present especially concentrated near the base of the unit. 250-400 2.84 Mangamate Tephra (MT), Strong brown and greyish brown and grey fine-med pumice and lithic Poutu Lapilli lapilli, shower-bedded and ungraded. 400 MT, Wharepu Tephra Upper 200 mm grey and olive grey med-coarse pumice ash shower- bedded, next 150 mm fine pumice lapilii of the same colour. Lower 50 mm strong brown fine pumice lapilii and coarse ash. 150 MT, Ohinepango Tephra? Strong brown and grey pumice and lithic lapilli and coarse ash. 150 Brown and yellow brown fine, firm, greasy ash. 150 Grey and strong brown pumice and lithic med ash. 150 Grey and strong brown fine pumice and lithic lapilli and coarse ash. 10 Brown fine ash. 1000 4.85 MT, Waihohonu Lapilii Alternating bands of grey and strong brown fine pumice and lithic lapilli and coarse ash. Shower-bedded but ungraded. Distinctive colour

A 41 banding. 600-1000 R04 Grey and greyish brown finesandy matrix diamicton. Pebble-cobble angular clasts, very clast rich almost clast supported. Unbedded, grey lithic clasts with occasional red and yellow clasts. 20 Yellow brown coarse pumice ash. 30 Grey lithic coarse ash. 300 MT, Oturere Lapilli Reversely graded greyish brown , strong brown and grey fine-med pumice and lithic lapilli. 30 6.23 94.9, Karapiti Tephra Olive grey fine-med ash containing thin (5 mm) white fine pumice ash. 4000+ R05-R09 Stacked sequence of diamicton units pinching in and out. Grey and greyish brown fine-coarse sandy matrices supporting clasts of pebble to boulders. Some of units have a horizontal fabric some are massive and unbedded. Lower units are very clast rich with huge boulders. 3000+ Yellow brown matrix diamicton, silt and sand matrix with pebble to large boulder clasts, grey clasts massive and unbedded. 13.23 Base into stream.

Section 32 Mangatoetoenui Stream, T20 445145 True left bank of stream 1500 m upstream of State Highway 1. Description taken from below Papakai Fonnation.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

200 c. 2.0 Mangamate Tephra (MT), Strong brown and grey fine-med pumice and lithic lapilli Poutu Lapilli 1200 MT, Wharepu Tephra Brownish grey and grey fine pumice and lithic lapilli and coarse ash. Shower-bedded but ungraded, lower 100 mm strong brown fine pumice and lithic fine lapilli. 20 Yellow brown finn, fine ash. 1000-2000 R03 Greyish brown diamicton. Fine-coarse sand matrix supporting clasts of pebble to boulder in size. 10 Yellow fine pumice lapilli. 30 Grey and strong brown coarse pumice and lithic ash. 200 Grey and strong brown mixed fine-med pumice and lithic lapilli. 100 5.56 MT, Waihohonu Lapilli Alternating layers of grey and strong brown finepum ice and lithic lapilli. 3000-4000 R04-R09 Several diamicton units, greyish brown and grey sandy matrix supporting pebble to boulder clasts. Dominantly grey clasts, some units have planar fabric others are massive. Lower unit contains very large boulder clasts in a yellow brown and greyish brown silty and sandy matrix. 2000+ Grey andesite lava flow 11.56 Base into stream.

Section 34 Mangatoetoenui Stream, T20 476160 True right bank of stream, next to the four-wheel-drive track and ford across the stream, 2 km downstream of S.H. 1.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

400 c. 1.0 Mangatawai Tephra Brown and dark brown fine ash with 20-30 mm beds of black and purplish grey med-fine ash. 100 Papakai Fonnation (Pp) Yellow brown fine greasy ash with paleosol development. 30 1.13 Waimahia Tephra within White fine ash cream-cakes within yellow brown fine ash. Pp 150 Pp Olive brown and yellow brown greasy fine ash. 100 1.38 Hinemaiaia Tephra within White med-coarse pumice ash scattered within fineyellow brown ash. Pp 80 Pp Yellow brown fine greasy ash. 500 R01 Yellow brown , greyish brown and grey sandy matrix diamicton. Grey and few red scoriaceous pebble to small boulder clasts supported within the matrix. 150 Pp Yellow brown fine ash with scattered yellow brown pumice and grey lithic fine lapilli. 40 2.15 Motutere Tephra within Pp Fine pale brown and white pocketing pumice ash, almost a complete layer, within brown fine ash. 100 Pp Greyish brown fine ash with scattered grey finelithic lapilli. 200-400 2.65 Mangamate Tephra (MT), Strong brown, olive brown and grey fine pumice and lithic lapilli, Poutu Lapilli shower-bedded but ungraded. 20 Grey coarse-med ash. 500 MT, Wharepu Tephra Greyish brown and grey fine pumice lapilli and coarse ash, shower bedded. Lower 150 mm strong brown fine pumice lapilli and coarse ash.

A 42 20 Yellow brown fine firm ash. 3000+ R03 Grey and pale olive grey sandy matrix diamictons. Pebble-boulder clasts supported within the matrix, mostly grey clasts, 2-3 units. 6. 19 Base into stream.

Section 35 Mangatoetoenui Stream, T20 476161 True left bank of stream, at 4WD ford in stream, 2 km downstream from State Highway 1.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

400 c. 1.0 Mangatawai Tephra Brown and dark brown fine ash with 20-30 mm thick layers of black and purplish grey fine-med ash. 50 Papakai Formation (Pp) Yellow brown fine greasy ash. 20 1.07 Waimahia Tephra within White fine pumice ash cream-cakes within fine yellow brown fine ash. Pp 50 Pp Grey fine-med ash mixed within yellow brown fine ash. 100 Yellow brown fine ash. 50 1.27 Hinemaiaia Tephra within White med-coarse ash scattered within yellow brown fine ash. Pp 150 Pp Yellow brown fine greasy ash. 40 1.46 Motutere Tephra within Pp Pale brown fine pumice ash cream-cakes within brown fineash. 100 Pp Greyish brown fine ash. 100-300 1.86 Mangamate Tephra (MT), Variable thickness strong brown and grey fine-med pumice lapilli. Poutu Lapilli 400 MT, Wharepu Tephra Upper 250 mm greyish brown and grey fine pumice and lithic lapilli and coarse ash, shower-bedded. Lower 150 mm strong brown finepumic e lapilli and coarse ash. 20 Yellow brown fine ash. Angular unconformity with sediments below. 0-1 000 R03 Grey and brownish grey sandy matrix diamicton. Pebble-boulder grey lithic clasts supported by the matrix. 100 3.38 MT, Ohinepango Tephra Coarse grey and strong brown ash, shower-bedded with alternating bands of colour. 800 MT, Waihohonu Lapilli Strong brown yellow brown and grey fine-med pumice and lithic lapilli and coarse ash, shower-bedded in bands of colour. 500 R04 Grey brownish grey and grey sandy matrix diamicton, faint planar fabric observed, clasts sparse and pebble-small boulders supported within the matrix. 1000+ R05-R06 Grey sandy matrix diamictons (2 units) massive and unbedded pebble- small boulder clasts, grey lithic. 5.68 Base onto track.

Section 37 Mangatoetoenui Stream, T20 465150 True left bank of stream 300 m downstream of State Highway 1, description from Poutu Lapilli.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

300 c. 2.0 Mangamate Tephra (MT), Strong brown and grey fine-med pumice and lithic lapilli, shower- Poutu Lapilli bedded. 30 Grey lithic med ash. 1000 3.03 MT, Wharepu Tephra Brownish grey and grey fine-med pumice and lithic lapilli and coarse ash, strongly shower-bedded and ungraded. Basal 100 mm strong brown fine pumice lapilli. 20 Yellow brown fine, firm ash. 30 Grey med lithic ash. 800 3.88 MT, Waihohonu Lapilli Grey, olive brown and strong brown fine-med pumice and lithic lapilli, shower bedded and ungraded. 50 Banded olive grey, brown and grey coarse-med pumice and lithic ash. 500 4.43 MT, Oturere Lapilli Grey, brownish grey and strong brown fine-med pumice and lithic lapilli, shower-bedded. 100 Pale grey and pale brown fine ash with scattered soft pumice and lithic fine lapilli. 30 Greyish brown med ash and scattered finelithic lapilli. 60 Upper half yellow brown fine-med ash, lower half yellow brown fine, soft pumice lapilli. 1000-1500 6.12 Pahoka Tephra Over-thickened at this locality. Grey and olive grey platy coarse ash and fine lapilli, normally graded top. Lower 500-1000 mm olive grey, and pale yellow med-coarse pumice lapilli, with banded lapilli common. 200 Yellow brown fine ash. 300-500 Distinctive very pale yellow and grey pumice and lithic coarse ash and fine lapilli. 400 7.22 Bullet Formation, Pourahu Pale yellow and pink fine ash, containing large pumice clasts of pale

A 43 Member yellow and pale olive, some clasts up to 250 mm diameter. 30 Pale yellow coarse-med pumice lapilli. 9000+ R07-R09 Grey, greyish brown and yellow brown matrix diamictons, several sandy matrix units. Pebble-boulder lithic clasts supported within matrix. 16.25 Base obscured.

Section 38 Ohinepango Stream, T20 4481 64 True left bank of stream close to foot bridge across stream along track to Waihohonu Hut.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

600 Makahikatoa Sand and Brown and yellow brown fine-med sands with interbedded black and Tufa Trig Formation grey faint med-fine 5-10 mm ash layers. 200 0.80 Taupo lgnimbrite White fine-coarse pumice ash with pumice lapilli and blocks intermixed. Occasional charcoal fragments. 400 Mangatawai Tephra Dark brown and dark greyish brown fine greasy ash with soil development. Interbedded black and purplish black fine-med ash, some containing yellow brown beech leaves. 50 Papakai Formation (Pp) Greyish brown greasy fine ash. 10 1.26 Waimahia Tephra within White pocketing fine ash within greyish brown fine ash. Pp 400 Pp Greyish brown and yellow brown fine greasy ash, with scattered occasional yellow brown and grey pumice and lithic fine iapilli. 50 1.71 Hinemaiaia Tephra within White med-coarse pumice ash scattered within greyish brown fine ash. Pp 300 Pp Yellow brown and greyish brown fine ash. 10 2.02 Motutere Tephra within Pp Pale brown fine pumice ash within yellow brown fine ash. 150 Pp Greyish brown and yellow brown fine ash with scattered finegrey lithic iapilli. 600 2.77 Mangamate Tephra (MT), Grey, olive grey and strong brown fine-med pumice and lithic lapilli, Poutu Lapilli shower-bedded and normally graded. 10 Fine grey ash. 1200 MT, Wharepu Tephra Olive grey and brownish grey fine-med pumice and lithic lapilii and coarse ash shower-bedded and ungraded. Lower 100 mm strong brown fine pumice lapilli. 20 Yellow brown fine, firm ash. 600 Reworked grey and strong brown med-coarse ash, planar bedded 20 mm scale. 50 MT, Waihohonu Lapilli Grey and strong brown coarse pumice and lithic ash. 400 5.05 Olive grey, strong brown and brownish grey fine pumice and lithic lapilli and coarse ash, shower-bedded and reverseiy graded, banded strong brown and grey zones near base. 100 R04 Pale brownish grey silly and sandy matrix diamicton. Grey pebble clasts supported within matrix. 10 Yellow brown fine pumice lapilli. 10 Grey fine ash. 20 Grey coarse lithic ash. 30 Pale brown and yellow fine ash with scattered fine pumice lapilli. 20 Olive grey fine ash. 400 5.64 MT, Oturere Lapilli Grey, brownish grey and strong brown fine-med pumice and lithic lapilli, shower-bedded but ungraded. 50 Pale yellow and grey fine-med ash grading into coarse pumice ash and lapilli of the same colours. 30 Yellow brown fine ash. 30 Yellow brown fine-med pumice iapilli with scattered grey fine lithic lapilli. 30 Greyish brown fine ash and finelithic lapilli. 300 6.08 Pahoka Tephra Upper 50 mm normally graded, olive grey platy med-coarse ash and fine lapilli. Lower 250 mm coarse-med pumice lapilli, olive grey and pale yellow with occasional banded pumice lapilli. 20 Yellow and grey coarse pumice and lithic ash. 250 R06 Grey sandy matrix diamicton. Grey pebble-cobble ciasts supported within matrix, eroded lower contact. 20 Dark yellow brown fine ash. 100 Yellow brown and greyish brown fine pumice lapilli. 100 Yellow brown fine pumice ash. 20 Grey and greyish brown med lithic ash. 200 6.79 Bullot Formation, Pourahu Poorly sorted yellow brown and pale brown fine-coarse lapilli. Lower Member 150 mm normally graded coarse pumice ash and fine lapilli of same colours, with scattered grey lithic coarse ash. 1500 R07 Grey and greyish brown sandy matrix diamicton. Pebble-boulder clasts supported within the matrix, grey, unbedded.

A 44 2500+ 93.16a Grey andesite lava flow 10.79 Base into stream

Section 40 Waihohonu Stream, T20 415815 Moraine capped ridge behind Waihohonu Hut, at top of un-vegetated part of hill. Description taken from Oturere Lapilli.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

500 c. 3.0 Mangamate Tephra, Greyish brown and grey fine-med pumice and lithic lapilli, shower- Oturere Lapilli bedded and ungraded. 100 Yellow brown fine ash with scattered pale yellow and yellow brown fine pumice lapilli. 100 3.20 Pahoka Tephra Pale olive grey and yellow fine pumice lapilli and coarse ash. Normally graded and platy coarse ash at top. Abundant yellow pumice. 10 Grey fine ash. 50 Brownish grey coarse pumiceash and fine pumice lapilli. 50 Greyish brown fine ash with scattered fine yellow pumice lapilli. 30 3.34 Bullot Formation, Pourahu Yellow brown med-fine pumice lapilli. Member? 300 Greyish brown and grey bedded sands and pebbles. 100 Yellow, strong brown and grey med-coarse pumice and lithic lapilli. 30 Yellow brown and greyish brown med-coarse pumiceash. 100 Pale brown, yellow and greyish brown fine pumice lapilli with scattered fine grey lithic lapilli. 30 Grey and pale yellow fine pumiceand lithic lapilli and coarse ash. 50 Grey, pale brown, and brownish grey fine-med pumice and lithic lapilli. 10 3.96 Waiohau Tephra White fine pumice ash cream-cakes. 100 Black and greyish brown coarse lithic ash and fine lapilli. 500+ Grey and greyish brown sandy matrix diamicton or till. Grey pebble- boulder clasts. 4.56 Base obscured

Section 41 Ohinepango Stream, T20 461170 True right bank of stream in a deep rut carved by a small tributary stream. 50 m west of State Highway 1.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

400 Taupo lgnlmbrite White fine-coarse pumice ash with intermixed pumice lapilli and blocks. Charcoal fragments contained near base of unit. 500 Mangatawai Tephra Brown and dark brown fine ash with paleosol development, Lower half has several interbedded black and purplish grey 20-30 mm, fine-med ash beds. Yellow brown beech leaves contained within some of the layers. 50 Papakai Formation (Pp) Brown fine greasy ash. 10 0.96 Waimahia Tephra within Pocketing white fine pumice ash within brown fine ash. Pp 200 Pp Brown and yellow brown fine greasy ash. 60 1.22 Hinemaiaia Tephra within White med-coarse ash scattered within yellow brown fine ash. Pp 250 Pp Yellow brown fine ash with scattered fineyellow pumice lapilli. 10 1.48 Motutere Tephra within Pp Pale brown fine pumice ash within yellow brown fine ash. 150 Pp Yellow brown fine ash with scattered grey pumice and lithic lapilli. 400 2.03 Mangamate Tephra (MT), Grey, strong brown and olive brown fine-med pumiceand lithic lapilli, Poutu Lapilli shower-bedded. 20 Grey med lithic ash. 400 MT. Wharepu Tephra Grey and brownish grey fine pumice and lithic lapilli, shower-bedded and ungraded. Lower 150 mm strong brown fine pumice lapilli. 10 Yellow brown fine ash. 100 MT, Ohinepango Tephra Alternating beds of grey and strong brown coarse pumice and lithic ash. 10 Grey fine ash. 80 MT, Waihohonu Lapilli Grey and strong brown fine pumiceand lithic lapilli, alternating colour layers. 20 Black and dark grey coarse ash. 20 Pale brown and pale yellow fineash with soft fine-med pumice lapilli. 20 Olive grey fine-med ash and fine lapilli, pumice and lithic. 500 3.21 MT, Oturere Lapilli Grey, brownish grey and strong brown fine-med pumice lapilli, shower bedded but ungraded. 50 Pale brown fine ash. 50-100 Pale yellow and grey coarse ash, pumice and lithic. 30 Pale brown fine ash.

A 45 30 Yellow and yellow brown soft pumice lapilli. 300 3.72 Pahoka Tephra Pale olive grey and pale yellow fine-coarse pumice lapilli, shower bedded and normally graded. Platy coarse ash near top. 20 Pale yellow and grey coarse pumice and lithic ash. 300-500 R06 Yellow brown fine sandy diamicton, with intermixed large pumice blocks 200 mm diameter, as well as lithic pebbles and cobbles supported within the matrix. 100 Yellow brown and strong brown fine pumice lapilli and coarse ash, shower-bedded. 30 Yellow brown fine ash with scattered fine pumice lapilli. 100 4.47 Bullot Formation, Pourahu Pale yellow and yellow brown fine-coarse pumice lapilli, with scattered Member grey fine lithic lapilli. 3000 R07 Clast rich diamicton, large boulder-pebble clasts with small amount of sandy and pebbly matrix. Grey, greyish brown and occasional red clasts and matrix. Erosive base, unconformity. 600 Bedded sands, silts and gravels planar and cross-bedded, grey, greyish brown and strong brown. 40 8.11 93.17, Kawakawa Tephra White fine pumice ash shower-bedded. 400 Grey and brownish grey bedded sands. 1000 9.51 R10 Yellow brown and pale yellow greasy silt matrix diamicton, multi- coloured pebble-large boulder, lithic and pumice clasts, highly weathered or altered clasts. 1000 R10 Strong brown and reddish brown silty matrix diamicton, with abundant reddish and grey lava bomb clasts within it. Semi-detached bomb clasts. 450 Bedded fine sands and silts, cross-bedded. Brown and brownish grey with scattered yellow fine pumice lapilli. 80 11.04 Yellow brown and strong brown med-coarse pumice lapilli. 500 Grey and olive grey bedded sands with scattered yellow brown fine pumice lapilli. 100 Strong brown and yellow brown fine-med pumice lapilli with scattered grey lithic lapilli. 1000+ R1 1 Greyish brown and grey silt and sand matrix diamicton, with pebble- boulder, grey lithic clasts supported within the matrix. 12.64 Base into stream.

Section 42 Ohinepango Stream, T20 4621 72 True right bank of stream, 50 m upstream from joining with Waihohonu Stream, immediately west of State Highway 1. Description taken from below Papakai Formation.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

200 c. 2.0 Mangamate Tephra (MT), Strong brown and grey pumiceand lithic fine-med lapilli, shower- Poutu Lapilli bedded, eroded upper contact. 500 MT, Wharepu Tephra Grey and brownish grey shower-bedded and ungraded fine-med pumice and lithic iapilli and coarse ash. Lower 200 mm strong brown fine pumice lapilli. 20 Yellow brown fine ash. 650 3.17 MT, Waihohonu Lapilli Grey, strong brown , and olive grey fine pumice and lithic lapilli, shower-bedded. Lower part has alternating strong brown and grey dominated bands. 10 Greyish brown and pale grey fine pumice lapilli. 10 Brownish grey med-coarse ash. 450 3.64 MT, Oturere Lapilli Grey, olive brown and yellow brown fine-med pumice lapilli, shower- bedded but ungraded. 80 Yellow brown and pink fine ash with interbedded med-coarse yellow and pale brown soft pumice lapilli. 40 Pink, grey and pale yellow med-coarse pumice ash. 120 3.88 Pahoka Tephra Grey and pale yellow fine pumice and lithic lapilli and coarse ash. 50 Pale yellow fine ash and fine-med pumice lapilli. 2000-3000 Grey andesitic lava flow. 2000 R10 Strong brown silty matrix diamicton, large scoriaceous bombs-pebbles supported within the matrix, grey and reddish coloured clasts. 50 Strong brown fine pumice lapilli and coarse ash. 1000+ Brown and brownish grey bedded silts and fine sands with layers and lenses of multicoloured pumice pebbles. 9.98 Base obscured

A 46 Section Waihohonu44 Stream, T20 464 174 Large road cutting on the western side of State Highway 1, immediately north of the bridge over the Waihohonu Stream.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

400 Ngauruhoe Formation and Dark brownish grey fine-med, friable ash with indistinct 10 mm thick Tufa Trig Formation grey and black fine-med ash beds. 400-1000 1.40 Taupo lgnimbrite White fine-coarse pumice ash with intermixed pumice lapilli and blocks, charcoal fragments also present. 450-500 Mangatawai Tephra Dark brown and greyish brown fine greasy ash with paleosol development, lower half of unit contains several black and purplish grey fine-medash layers, often with preserved yellow brown beech leaves. 200 Papakai Formation (Pp) Brown and greyish brown fine ashwith paleosol development. 50 Greyish brown fineash with pocketing black fine-med ash. 20 2.17 Waimahia Tephra within Cream-cakes of fine white pumice ash within brown finegreasy ash. Pp 100 Pp Yellow brown fine greasy ash with scattered fine yellow brown soft pumice lapilli. 50 2.32 Hinemaiaia Tephra within White med-coarse pumice ash scattered within yellow brown fineash. Pp 650 Pp Yellow brown greasy fine ash wi th scattered yellow and grey fine pumice and lithic fine lapilli, more concentrated near the base. Uneven lower contact. 350 Reworked strong brown and grey pumice and lithic fine lapilli and coarse ash, eroded upper contact, planar and cross-bedded. 400 3.42 Mangamate Tephra (MT). Strong brown and grey fine pumice and lithic lapilli, shower-bedded but Poutu Lapilli ungraded. 600 MT, Wharepu Tephra Upper 450 mm, grey, greyish brown and some strong brown fine-med pumice lapilli, shower-bedded but ungraded. Lower 150 mm strong brown fine pumice lapilli. 20 Yellow brown fine, firm ash. 30 MT, Ohinepango Tephra Greyish brown and strong brown med-coarse pumice and lithic ash. 30 Greyish brown reworked fine-med pumice and lithic lapilli, within med grey and greyish brown ash. 1000 5.10 MT, Waihohonu Lapilli Strong brown, grey and greyish brown fine-med pumice and lithic lapilli and coarse ash with strong shower-bedding giving a banded colour appearance. 150 Greyish brown fine-med sand, with mixed grey sub-rounded pebbles, and occasional cobbles. 20 Strong brown and brown coarse ash and fine pumice lapilli. 450 5.72 MT, Oturere Member Grey, greyish brown and strong brown med-fine pumice and lithic lapilli, shower-bedded but ungraded. 60 Yellow brown and pale brown fine ash with scattered grey-white soft pumice fine lapilli. 20 Yellow brown and strong brown fine pumice lapilli and coarse ash. 100-150 Grey and greyish brown sands and finegravels planar bedded, sub- rounded cfasts. 10 Strong brown fine-med pumice lapilli. 30 Greyish brown sand and fine, sub-rounded gravel. 350 6.34 Pahoka Tephra Olive grey and pale yellow fine-med pumice lapilli and coarse ash, normally graded, upper part has very platy lapilli, and lower part contains occasional banded olive grey and pale yellow pumice lapilli. 20 Strong brown and yellow brown coarse pumice ash. 1200 Grey and greyish brown bedded sands and fine-coarse gravels, planar bedded 100-300 mm scale, sub-rounded clasts. 600-1 300 8.86 R06 Greyish brown sandy matrix diamicton, abundant boulder-cobble clasts in upper half of unit and dominantly pebble clasts within the lower half, clasts supported within the matrix. Dominantly grey with occasional red lithic clasts. 50 Pale brown fine ash. 100 Pale brown fine-med pumice lapilli. 100 9.11 Bullet Formation, Pourahu Pale brown and pinkish fine ash containing abundant coarse-med Member pumice lapilli within. 100-200 Pale yellow and pale brown fine-med pumice lapilli and coarse ash with scattered grey coarse lithic ash. 10 Brown fine silty layer. 3000+ R07 Greyish brown sandy and silty matrix diamicton. Pebble-large boulder cfasts supported within the matrix, dominantly grey with occasional red clasts. Diamicton pinches out over lava, laps up onto side of lava. Angular unconformity with coverbeds below diamicton. 2000+ R09ff3 Brownish grey sand and silt matrix diamicton, dominantly multi- coloured and grey pumice and lithic pebble clasts with few grey lithic

A 47 boulders, unconformable upper contact. 400 14.72 Kawakawa Tephra Shower-bedded white and pale pink fine-med pumice ash. Upper 150 mm massive fine pale pink ash, onto 100 mm pale brown med ash with coarse ash horizons and accretionary lapilli present up to 15 mm in diameter. Onto 30 mm coarse white pumice ash, onto 20 mm pale brown fine ash, onto 30 mm fine white ash. 200 Dark greyish brown fine-med ash. 50 Yellow brown fine-coarse pumice lapilli intermixed within the scoriaceous top of the lava flow. 30000+ 93.19 Grey andesite lava flow with red scoriaceous top, highly brecciated top. 44.97 Base into stream on opposite side of road.

Section 45 Upper Waikato Stream, T20 461096 True left bank of northern tributary 500 m upstream of State Highway 1. Description taken from M1 marker sequence in Bullot Formation.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

10 c. 4.0 Bullot formation, M1 Grey fine ash. 10 Yellow brown coarse pumice ash. 20 Yellow brown fine ash with scattered grey coarse lapilli. 30 Grey fine lithic lapilli and coarse ash. 30 Yellow brown coarse pumice ash and fine iapilli. 20 Grey coarse lithic ash. 40 Grey and yellow brown coarse ash and finepumice and lithic lapilli. 150 Grey and strong brown fine-med pumice and lithic lapilli. 50 Grey and purplish grey med ash. 200 4.55 Strong brown fine-med pumice lapilli, shower-bedded with scattered grey lithic fine lapilli. 30 Grey and strong brown lithic and pumice med ash. 20 Strong brown and grey fine-med pumice and lithic lapilli. 50 Grey and brownish grey coarse-med pumice and lithic ash. 100 4.75 Strong brown and brownish grey fine-med pumice lapilli with scattered grey fine lithic iapilli. 20 Brownish grey med ash. 20 Strong brown and yellow brown finepumice lapilli with scattered grey fine lithic lapilli. 20 Pale grey lithic med ash. 20 Strong brown fine-med pumice lapilli. 20 Purplish grey lithic coarse ash. 10 Brown coarse ash and finepumice lapilli. 40 Purplish grey coarse ash with scattered yellow brown fine pumice ash. 20 Strong brown and yellow brown finepumice lapilli with scattered fine grey lithic lapilli. 20 Brownish grey and grey coarse lithic and pumice ash. 30 Yellow brown and brown med pumice ash with sparse scattered grey fine lithic lapilli. 50 5.02 Waiohau Tephra? Brownish grey and grey coarse pumice and lithic ash, contains a 10-20 mm pocketing white fine pumice ash. 100 Grey and brownish grey coarse pumice and lithic ash with scattered yellow brown and strong brown fine pumice iapilli. 80 Strong brown fine-med pumice iapilli, shower-bedded with scattered grey fine lithic lapilli. 20 Black lithic med-coarse ash. 400 5.64 Hinuera Formation Bedded silt, sand and gravels, wavy and planar bedding 20-100 mm scale. intermixed white fine pumice ash in places. Greyish brown and pale brown. 600 Brownish grey sand and finegravel. Planar and wavy bedding on a 10-20 mm scale, unconsolidated with pumice and lithic pebbles. 500 6.74 R10 Eroded upper contact. Strong brown sand and silt matrix diamicton. Multi-coloured pebble and cobble clasts supported by the matrix. Clasts highly weathered lithic and pumice, heterolithologic. 100 Yellow brown silt matrix supporting clasts as described in the unit above. 1000 Greyish brown massive sandy matrix diamicton unit wi th fine pebbly clasts, dominated by grey lithic and pumice clasts with multi coloured pebbles also. 600 8.44 Yellow brown and fine sand and silt matrix diamicton with multi coloured cobble and boulder clasts. 300 Bedded brown and greyish brown sands and gravels, wavy and planar bedded. 200 Pale brown and brown fine sands and silts 10-20 mm scale wavy and

A 48 planar bedding. 200 Grey and brown sands and gravels cross bedded. 700 R10 Pale brown and pale yellow brown silt matrix diamicton with cobble and pebble clasts supported within the matrix. 3000 12.84 Bedded fluvial sands, gravels and silt layers, grey lithic and pumice pebbles. Well developed cross-stratification in some layers. Lenses of yellow pumice lapilli interbedded in places, and some layers slightly hardened. 250 12.99 93.23 Yellow brown pumice fine-med lapilli with few scattered fine grey lithic lapilli. 2000+ Bedded silt, sand, and gravel in lenses and layers. Yellow brown and brown silts with grey and greyish brown sands and gravels. 14.99 Base into stream.

Section 46 Upper Waikato Stream, T20 464098 True left bank of northern tributary of stream, 300 m upstream from State Highway 1. Description taken from lowest Bullet Formation tephra preserved.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

80 c. 6.0 Yellow brown and reddish brown fine-med pumice lapilli with scattered fine grey lithic lapilli. 20 Grey med-coarse ash. 300 Yellow brown , bedded fine med sand and silt, with occasional yellow brown fine pumice pebbles, weak planar beds. 500 Grey lithic, bedded sands and fine-coarse gravels, mostly grey lithic clasts with occasional red clast, cross-bedding, eroded lower contact. 50-200 7.02 R10 Variable top brown silt and sand matrix diamicton, containing multi coloured pumice and weathered lithic pebble clasts supported within the matrix. 200 Yellow brown silt and finesand matrix with fine pebbles-boulders supported within the matrix. 500 Massive pale brown silt and sand matrix diamicton with multi-coloured pumice and highly weathered lithic pebble-cobble clasts supported within the matrix. 2500 Strong brown, brownish grey and purplish grey bedded sands and gravels, well developed planar and cross-bedding on a 10-50 mm scale. Eroded lower contact. 0-50 10.27 93.24 Strong brown and yellow brown fine pumice lapilli, soft and porous. 200 Grey bedded sands and gravels as above. 200 93.25 Strong brown reversely graded coarse pumice ash-med lapilli, with scattered grey lithic fine lapilli. 10 Brown fine ash. 500 11.18 93.26, Marker Unit 1 Grey, olive grey and yellow coarse-med pumice and lithic ash, shower bedded with distinctive banded appearance. Lower 100 mm coarse ash and fine pumice and lithic lapilli. 20 11.20 93.27, Marker Unit 1 Yellow fine pumice iapilli and coarse ash. 400 Brown and greyish brown bedded sands and gravels, planar and cross-bedded with multi-coloured heterolithologic pumice and lithic pebbles in layers and lenses. 500+ R1 1 Massive yellow brown and grey sand and silt matrix diamicton, grey and greyish brown lithic pebble-cobble clasts supported by the matrix. 12.10 Base into stream.

Section 47 Upper Waikato Stream, T20 465099 True left bank of northern tributary, 1 OOm upstream from State Highway 1, description taken from below Marker Unit 1.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

2000 c. 10 Grey, and dark grey bedded sands and coarse gravels, wavy and planar bedding 200 mm scale beds. 1000 Grey and brownish grey bedded sands and fine gravels, cross-bedded and planar bedded on a 20-50 mm scale. 2000 Brown and strong brown bedded sands, silts and gravels, planar and cross-bedded with layers and lenses of multi-coloured pumice and lithic clasts, yellow pumice and grey lithic clasts dominant, 10-100 mm scale beds. 400 13.40 93.28 Yellow and yellow brown soft fine-med pumice lapilli, with sparse scattered grey lithic fine lapilli. 1500 Firm, bedded sands silts and gravels, grey and greyish brown planar and cross-bedded, 10-100 mm scale.

A 49 1000+ R1 1 Very hard grey and greyish brown sandy matrix diamicton. Grey pebble-boulder clasts supported wi thin the matrix. Faint planar fabric. 15.90 Base into stream.

Section 50 Tongariro River, T20 494965 Road cutting and cliff exposure on the true left bank of river at the end of Waipakihi Road near the river gauging station.

Unit Depth Cum. Correlation and samples Description (mm) Depth (m) taken

1500 Taupo lgnimbrite Fine-coarse white pumice ash with mixed pumice lapilli and blocks. Charcoal fragments contained within the lower part of the unit. 300 Mangatawai Tephra Dark brown and greyish brown fine ash with paleosol development. Interbedded within this are several 10-20 mm beds of fine-med black and purplish grey ash. Some ash beds contain Yellow brown beech leaves preserved within. 600 2.40 Papakai Formation Yellow brown and brown fine greasy ash with paleosol development and scattered yellow brown and grey fine pumice iapilli. 400-500 Mangamate Tephra (MT), Variable upper contact. Yellow brown and grey fine-med poorly Poutu Lapilli vesicular pumice and lithic lapilli, shower-bedded and ungraded. 10 Black med lithic ash. 300-350 3.26 MT, Wharepu Tephra Grey and greyish brown fine pumice and lithic lapilli and coarse ash, shower-bedded but ungraded. Basal 50 mm strong brown fine pumice. 10 Yellow brown fine, firm ash. 250-300 MT, Waihohonu Lapilli Grey and strong brown alternating coloured. Shower-bands of pumice and lithic fine lapilli and coarse ash. 10 Brown fine ash. 250 3.83 Grey and strong brown fine pumice and lithic fine lapilli and coarse ash, shower-bedded and normally graded. 20 Pale yellow fine pumice lapilli and coarse ash. 50 Grey fine lapilli and coarse lithic ash. 20 Pale yellow and grey pumice and lithic coarse ash and fine lapilli. 10 Grey coarse lithic ash. 20 Yellow brown and brown fineash with scattered fine yellow pumice lapilli. 20 Grey coarse-med lithic ash. 300 4.27 Mt, Oturere Lapilli Grey and pale yellow fine-med pumice and lithic lapilli, shower-bedded and ungraded. 50 Brown and yellow brown greasy fine ash. 200 Pahoka Tephra Grey and olive grey fine pumice lapilli and coarse ash, platy coarse ash at top of unit, normally graded. Occasional banded lapilli present, olive grey and pale yellow bands. 10 Pale brown coarse pumice ash. 10 Grey med-coarse lithic ash. 100-200 Yellow brown fine ash with scattered yellow fine pumice lapilli. 150-200 4.94 Bullot Formation, Pourahu Yellow brown and pale yellow fine-med pumice lapilli and coarse ash, Tephra normally graded. 50 Grey lithic fine lapilli. 1500-2000 R07 Grey and yellow brown sand and silt matrix diamicton. Upper 1000 mm grey pebble-boulder clasts, very clast rich and almost clast supported in places. Lower part finergrained with matrix mostly supporting pebble clasts, grey and olive grey lithic and sparse poorly vesicular pumice. Lower part has faint planar fabric. 800 R08-R09 Yellow brown silt and grey sand matrix diamicton, planar fabric preserved. Yellow, strong brown and grey fine pebbly pumice and lithic clasts supported by the matrix. 1000 8.74 Yellow brown and greyish brown silt and sand matrix. Large boulder- pebble clasts supported by the matrix, reversely graded, upper part dominated by boulders and lower part by cobbles-pebble clasts. 4000-5000 Grey and brownish grey sandy and silt matrix diamicton with large boulder-pebble clasts. Massive and unbedded. Lower 2000 mm finer grained with largest clasts dominantly grey cobbles. Sharp contact to unit below. 1000+ 14.70 Grey and greyish brown firm sandy matrix diamicton, grey angular lithic pebble-cobble clasts supported by the matrix. Faint planar fabric preserved. 5000 Obscured 2000+ 21 .70 Bedded and tilted greywacke rock, with base into river.

A 50 Section 51 Tongariro River, T20 496157 Large road cutting along the Waipakihi Road, after the crossing of the unnamed stream on the top of the large high surface.

Unit Depth Cum. Correlation and samples Description (mm) Depth (m) taken

300 Ngauruhoe Formation Brown greasy fine-med ash with present soil development. 200-1 000 Taupe lgnimbrite White fine-coarse pumice ash with pumice lapilli and blocks intermixed. Charcoal fragments intermixed near base of unit. 200-300 1.60 Mangatawai Tephra Dark brown and greyish brown fine ash with paleosol development, and several interbedded black and purplish grey fine-med ash beds, 10-20 mm thick, some of the beds contain yellow brown beech leaves preserved within. 80 Papakai Formation (Pp) Brown and yellow brown fine greasy ash with occasionalscattered fine yellow pumice lapilli. 20-30 1.71 Waimahia Tephra within Pocketing white fine ash within yellow brown fine ash. Pp 70 Pp Yellow brown fine greasy ash. 100 1.88 Hinemaiaia Tephra within White med-coarse pumice ash scattered within yellow brown greasy Pp fine ash. 250 Pp Yellow brown greasy fine ash with scattered fineyellow pumice lapilli. 20-30 2.16 Motutere Tephra within Pp Pale brown fine pumice ash cream-cakes. 100 Pp Yellow brown fine ash with scattered grey fine lithic and pumice lapilli. 400-600 2.86 Mangamate Tephra (MT), Wavy upper contact. Strong brown , grey and greyish brown fine Poutu Lapilli pumice and lithic lapilli, shower-bedded and weakly reversely graded. 10 Grey med lithic ash 300-400 MT. Wharepu Tephra Grey and greyish brown fine pumice and lithic lapilli and coarse ash, shower-bedded and ungraded. Lower 100 mm strong brown fine pumice lapilli. 10 Brown fine firmash. 20-50 3.33 Poronui Tephra Grey med-coarse ash containing pocketing finewhite pumice ash. 150 Alternating shower-bands of grey and strong brown med-coarse pumice and lithic ash. 200 3.68 MT, Waihohonu Lapilli Strong brown, grey and greyish brown alternating bands of fine pumice and lithic lapilli. 10 Brown fine ash. 1000 Grey and strong brown coarse ash and fine-med pumice lapilli, shower-bedded with bands of alternating colours, ungraded. 10 Grey fine ash. 10-20 Grey and yellow brown pumice and lithic med ash. 40 Yellow brown fineash with scattered grey and yellow brown fine pumice lapilli. 500 5.26 MT, Oturere Lapilli Grey, greyish brown and strong brown fine-med pumice and lithic lapilli, shower-bedded but ungraded. 50-100 Yellow brown fine ash with scattered fine yellow and yellow brown pumice lapilli. 20 Yellow brown fine pumice lapilli. 50 Yellow brown fineash. 400 5.83 Pahoka Tephra Normally graded coarse ash-med pumice lapilli, olive grey and pale brown , with scattered banded pumice lapilli with grey and pale brown stripes. Platy pumice coarse ash at the top of the unit. 10 Pale yellow and grey pumice and lithic coarse ash. 40 Yellow brown fine ash. 50 5.93 Bullet formation, Pourahu Yellow brown pumice lapilli and coarse ash. Member? 50 Yellow brown fine ash. 40 Strong brown and pale yellow fine-med pumice lapilli, with scattered grey fine lithic lapilli. 40 Grey and pale yellow coarse pumice and lithic ash. 30-40 Strong brown, pale yellow and yellow brown fine-med pumice lapilli. 50 6. 17 Pale brown and pale greyish brown fine ash with scattered fine grey lithic lapilli. 20 Pale yellow and yellow brown pumicelapilli wi th scattered grey lithic coarse ash. 20 Grey fine ash. 20 Yellow brown fine pumice lapilli with scattered finegrey lithic lapilli. 50 Grey and greyish brown med-coarse lithic and pumice lapilli. 10 Olive grey and brown fine ash. 40 Greyish brown and purplish grey coarse lithic ash and fine lapilli. 10 Pale yellow coarse pumice ash. 20 Grey coarse lithic ash. 30 6.39 Yellow brown fine-med pumice lapilli with scattered grey lithic fine lapilli. 20-50 Grey coarse lithic ash.

A 51 50 Pale yellow fine pumice lapilli, reversely graded with grey coarse lithic ash. 20 Grey coarse ash and fine lithic lapilli. 10 Pale greyish brown and yellow brown fineash with scattered yellow brown fine pumice lapilli. 300 Grey and pale yellow reworked pumice and lithic coarse ash and fine lithic lapilli, wavy and planar beds. 40 6.86 Strong brown fine pumice lapilli with scattered grey fine lithic lapilli. 30 Greyish brown and grey coarse-med ash. 100 Strong brown med-fine pumice lapilli with scattered grey fine lithic lapilli and med ash, shower-bedded and ungraded. 50-100 Purplish black coarse lithic ash and finelapilli. 50 Grey and greyish brown med-coarse lithic ash. 40 7.18 Waiohau Tephra Greyish brown med ash with interbedded white fine pumiceash. 10 Purplish grey med lithic ash. 300 Alternating layers of greyish brown and olive grey coarse lithic and pumice ash and fine lapilli. 30 Strong brown and pale yellow fine pumice lapilli. 200 7.72 Purplish black coarse lithic ash. 200 Greyish brown, grey and yellow brown fine-med shower-bedded pumice lapilli. 100 Pale yellow brown and brown fine-med ash with scattered coarse pumice ash. 50 Yellow and grey coarse pumice and lithic ash. 50 Brown and greyish brown fine ash. 30 Greyish brown and pale brown fine pumice lapilli with scattered grey lithic fine lapilli. 100 8.25 Normally graded grey and pale olive brown coarse pumice ash and fine lapilli. 50 Greyish brown and pale brown med-coarse pumice ash with scattered grey lithic med ash. 30 Pale yellow pumice fine lapilli with scattered grey lithic coarse ash. 30 Greyish brown fine ash. 10 Purplish grey fine ash. 150-200 8.57 Reversely graded yellow brown fine-med pumice lapilli with scattered fine grey lithic lapilli. 50 Pale yellow brown normally graded coarse pumiceash and finelapilli, with scattered grey lithic coarse ash. 0-1 000 9.62 R10 Brown silt and sand matrix diamicton, sparse grey angular lithic pebble-large boulder clasts supported by the matrix, angular unconformable lower contact. 500-1000 Brownish grey and brown sand and silt matrix firm diamicton. Weak planar fabric preserved and pebble clasts supported by the matrix. Grey lithic and yellow pumice clasts. 400 Greyish brown bedded sands and gravels, cross-bedded, sub-rounded grey lithic gravels. 1000 12.02 Massive firm yellow brown silt and sand matrix diamicton. Pebble cobble clasts supported by the matrix, grey lithic and abundant yellow pumice clasts. At top of unit abundant yellow pumice pebbles with a faint horizontal fabric, the rest of the unit is massive and unbedded. 20-30 12.05 94.13 Yellow brown and yellow coarse pumice ash with scattered grey lithic coarse ash. 40 Brown fine ash. 3000 R1 1 Massive firm fine-grained diamicton. Brown silt and sand matrix supporting clasts of grey and red lithic and yellow pumice pebble clasts. 15.09 Base obscured.

Section 53 Waihohonu Stream, T20 4981 84 True right bank of stream, 300 m upstream of where Pangara Stream meets Waihohonu Stream.

Unit Depth Cum. Correlation and samples Description (mm) Depth (m) taken

1000- Taupe ignimbrite White and pale grey fine-coarse ash with abundant pumice blocks and 10000 lapilli intermixed. Large charcoalised logs preserved especially near the base of the unit. Eroded angular contactint o units below, angular unconformity. 200-350 10.35 Mangamate Tephra (MT), Strong brown, greyish brown and grey coarse pumice and lithic ash Waihohonu lapilli and fine lapilli, shower-bedded with alternating coloured zones. 20 Grey and greyish brown fine-med ash. 20 Yellow brown and strong brown fine pumice iapilli and coarse ash. 30 Greyish brown med-coarse pumice ash. 450 10.87 MT, Oturere Lapilli Reddish brown, and grey fine-med poorly vesicular pumice and lithic

A 52 fine-med lapilli, shower-bedded but ungraded. 50 Fine yellow brown and greyish brown ash with scattered grey and yellow brown fine pumice and lithic lapilli. 50 Yellow brown, yellow and grey coarse pumice and lithic ash. 100 Yellow brown fineash with scattered grey and yellow brown lithic and fine pumice lapilli. 200 11.27 Pahoka Tephra Olive grey and pale brown platy coarse pumice ash and fine-med lapilli. Normally graded, with occasional olive grey and pale brown banded pumice lapilli, shower-bedded. 40 Brown and yellow brown fine ash. 150 Grey med-coarse sands with scattered yellow brown fine pumice lapilli, planar bedded on a 10 mm scale. 3000 Bedded greyish brown and grey sands and fine-coarse gravels. Planar bedded on a 100 mm scale. 1300 15.76 Bedded grey sands with scattered common pale brown pumice fine pebbles. Wavy bedded with occasional lenses of pumice pebbles. 40 Greyish brown and pale brown finesand and silt, wavy bedded. 100-150 15.95 94. 18 Strong brown fine-med pumice lapilli, interbedded within top of diamicton and as a discrete layer on top of diamicton. 5000- R07-R09 Clast rich diamicton unit, dominating exposures all along this stream. 10000 Grey and occasional red lithic sub-rounded clasts, pebble-very large boulders. Matrix of greyish brown silt and fine sand, yellow pumice and grey lithic pebble clasts common. Matrix supported but very clast rich. 2000+ 27.95 Tongariro andesite Grey lava, base into stream

Section Oturere Strea54 m, T19 484209 Cutting on the western side of State Highway 1, 100 m south of crossing of Oturere stream. True left bank of Oturere stream valley.

Unit Depth Cum. Correlation and samples Description (mm) Depth (m) taken

600-1000 Taupo lgnimbrite White pumice, fine-coarse ash with mixed pumice lapilli and blocks, charcoal fragments preserved within base of unit. 400-500 Mangatawai Tephra Dark brown and dark greyish brown fine ash with paleosol development. Interbedded black and purplish grey fine-med ash beds 10-20 mm thick with common pale brown beech leaves preserved within layers. 100 1.60 Papakai Formation (Pp) Brown greasy, fine ash with paleosol development. 20 1.62 Waimahia Tephra within Cream-cakes of white, fine, pumice ash within brown greasy fine ash. Pp 150 Pp Dark greyish brown greasy fineash. 100 1.87 Hinemaiaia Tephra within White, med-coarse pumice ash scattered within fine greasy greyish Pp brown ash. 150 Pp Yellow brown and brown greasy fine ash with vertical cracking, and paleosol development. 10-20 2.04 Motutere Tephra within Pp Pale brown pumice fine ash cream-cakes within brown fine greasy ash. 100 Pp Yellow brown and brown fineash. 550 2.69 Mangamate Tephra (MT), Strong brown and grey fine-med poorly vesicular pumiceand lithic Poutu Lapilli lapilli, shower-bedded but ungraded. 40 Grey coarse-med lithic ash. 50 MT, Wharepu Tephra Strong brown fine-med pumice lapilli. 150 Grey and strong brown 10 mm alternating layers of poorly vesicular pumice and lithic med-coarse ash. Reworked ash. 600 MT, Waihohonu Lapilli Strong brown , greyish brown and grey fine-med pumice and lithic lapilli. Shower-bedded but ungraded. Lower 40 mm yellow brown pumice and lithic lapilli. 10 Greyish brown fine ash. 10 Grey med lithic ash. 10 Yellow brown med-coarse pumice ash. 10 Grey fine ash. 400 3.97 MT, Oturere Lapilli Grey and greyish brown fine-med poorly vesicular pumice and lithic lapilli shower-bedded but ungraded. 150 Pale brown soft greasy fine ash with interbedded yellow brown and grey fine lapilli. 150 3.27 Pahoka Tephra Grey, olive grey and pale brown fine-med pumice lapilli, shower bedded, platy fine lapilli near top. 50 Brown fine ash, soft and greasy with scattered yellow brown fine pumice lapilli. 50 Pale yellow and grey coarse pumice and lithic ash. 100 Brown and yellow brown fine ash with scattered fine, soft, brown and yellow fine pumice lapilli.

A 53 50-100 3.57 Bullot Formation, Pourahu Yellow brown and pale brown fine-med pumice lapilli with scattered Member grey lithic lapilli, shower-bedded. 100 Grey and pale greyish brown fine ash with interbedded yellow brown and grey fine pumice and lithic lapilli. 40 Yellow brown fine-med soft pumice lapilli with scattered grey fine lithic lapilli. 180 Grey and greyish brown fine ash with scattered yellow brown and grey pumice and lithic lapilli. 20 Yellow brown fine pumicelapilli. 100 Brown and pale brown fine-med ash with scattered yellow brown fine pumice lapilli. 80 4.09 Yellow brown and strong brown fine-med pumice lapilli with scattered grey lithic fine lapilli. 150 Yellow brown coarse pumice ash and fine lapilli wi th sparse, scattered grey lithic coarse ash. 100 Grey and few yellow brown coarse lithic and pumice ash and fine lapilli. 50 Brown and greyish brown finegreasy ash with scattered yellow brown and strong brown fine pumice lapilli. 20 4.41 Waiohau Tephra Grey and greyish brown fine ash with pocketing white fine pumice ash. 80 Greyish brown and brown fine-coarse ash with sparse scattered fine yellow brown pumice lapilli. 100 Yellow brown fine pumice lapilli, shower-bedded with scattered fine grey lithic lapilli. 50 Grey fine-med ash with scattered fine yellow brown pumice lapilli. 30 Yellow brown , strong brown and grey med-coarse pumiceand lithic ash. 30 Brown fine ash. 150 4.85 Pale brown and pale yellow pumicefine-med lapilli with scattered grey fine lithic lapilli, grey lithic rich zone in centre of unit. 40 Brown fine ash with scattered yellow brown fine pumice lapilli. 100 Yellow and pale yellow brown fine-med, soft pumice lapilli with scattered grey lithic coarse ash and fine lapilli. 10 Grey med lithic ash. 30 Yellow and pale yellow brown fine-med, soft pumice lapilli with scattered grey lithic coarse ash and fine lapilli. 10 Grey med lithic ash. 80 5.12 Yellow brown and pale brown fine pumice lapilli with scattered grey lithic fine lapilli. 180 Greyish brown and grey fine ash wi th scattered pale brown fine pumice lapilli. Pale brown and yellow brown fine pumice lapilli with scattered grey 20 lithic fine lapilli. 200 Brown and greyish brown fine ashwith scattered grey and pale brown lithic and pumice fine lapilli. 50-60 Grey and greyish brown coarse lithic and pumice ash and fine lapilli. 50 Grey and greyish brown fine-med lithic ash. 50 5.68 Pale brown coarse pumice ash and fine lapilli, normally graded, with scattered grey lithic coarse ash and fine lapilli. 30 Yellow brown fineash. 40 Yellow brown and grey coarse pumice and lithic ash mixed in equal proportions. 100 Grey and greyish brown fine ash with scattered pale brown fine pumice lapilli. 20 Greyish brown and yellow brown fine pumice lapilli. 20 Purplish grey fine ash. 120 6.01 94.20 Greyish brown and pale brown med-fine pumice and scattered lithic lapilli. 350 Brown and greyish brown fine ash with scattered grey and yellow brown lithic and pumice fine lapilli. 400 Grey and greyish brown fine ash or silt with occasional interbedded cobbles and yellow brown and grey lithic fine lapilli. 2000-3000 T1 Reworked diamicton, grey clast supported with a grey sandy matrix, pebble cobble and small boulder clasts. 4000+ T2 Eroded upper contact. Yellow brown. pale brown and greyish brown silt and sand matrix diamicton. Pebble-large boulder clasts supported within matrix, massive with no bedding, multi-coloured pebble clasts, larger clasts dominantly grey. 13.76 Base obscured.

A 54 Section 56 Oturere Stream, T19 478213 True left bank of stream 500 m upstream of previous section.

Unit Depth Cum. Correlation and sample Description (mm) depth (m) taken

500-3000 Taupo lgnimbrite White and pale grey fine-coarse pumice ash with mixed pumice lapilli and blocks. Charcoal fragments preserved near base, thickens in paleo-valley. 800-1000 4.0 Mangatawai Tephra Dark brown fine ash with paleosol development, several purplish grey and black ash beds, 10-30 mm thick interbedded within. Common pale brown beech leaves preserved within black ash layers. 500 Papakai Formation (Pp) Yellow brown and dark brown fine greasy ash with paleosol development. 100 4.6 Hinemaiaia Tephra within White med-coarse pumice ash scattered within yellow brown fine Pp greasy ash. 200 Pp Yellow brown fine greasy ash. 10-20 4.82 Motutere Tephra within Pp Pale brown fine ash cream-cakeswithin yellow brown fine ash. 100 Pp Yellow brown fine ash. 200-500 5.42 Mangamate Tephra (MT), Uneven upper boundary. Strong brown and grey fine-med poorly Poutu Lapilli vesicular pumice and lithic lapilli, shower-bedded but ungraded. 40 Grey lithic med-coarse ash. 50 MT, Wharepu Tephra Strong brown fine pumice lapilli. 200-400 MT, Ohinepango Tephra Grey and strong brown alternating layers of coarse pumice and lithic ash, shower-bedded. 300 MT, Waihohonu Lapilli Grey, greyish brown and strong brown fine-med pumice and lithic lapilli, shower-bedded. 500 Grey and strong brown fine poorly vesicular pumice and lithic lapilli. Grey dominated at the top and strong brown at the base. 30 Grey lithic med ash. 20 Strong brown coarse pumice ash. 10 Grey coarse ash. 350-400 7.17 MT, Oturere Lapilli Grey, greyish brown and strong brown fine-med poorly vesicular pumice and lithic lapilli, shower-bedded and ungraded. 50 Greyish brown fine ash with scattered yellow brown and pale brown fine pumice lapilli. 20 Yellow brown coarse pumice ash. 20 Greyish brown fine ash. 30 Strong brown and yellow brown fine pumice lapilli. 200 7.49 Pahoka Tephra Grey and olive brown fine-med pumice lapilli, shower-bedded, with scattered banded grey and pale brown pumice lapilli. 10 Grey coarse lithic ash. 10 Yellow brown coarse pumice ash. 40 Yellow brown and brown fine ash. 80 Yellow brown and grey coarse pumice and lithic ash and fine lapilli, equal proportions of lithics and pumice. 40 Dark brown fine ash, with scattered yellow brown and brown fine pumice lapilli. 100 7.77 Bullot Formation, Pourahu Yellow brown , pale brown and strong brown med-fine pumice lapilli, Member with scattered fine grey lithic lapilli. 40 Greyish brown fine ash with scattered grey and yellow brown fine lithic and pumice fine lapilli. 40 Strong brown fine pumice lapilli. 30 Grey lithic coarse-med ash. 10 Yellow brown coarse pumice ash. 40 Grey med-coarse lithic ash. 20 Yellow brown and pale brown fine soft pumice lapilli with scattered grey lithic lapilli. 50 8.00 Greyish brown and grey pumice and lithic med ash. 30 Grey coarse lithic ash. 30 Yellow brown and brown fine pumice lapilli with scattered grey coarse ash. 100 Brown and yellow brown fine ash with scattered, sparse yellow brown and grey fine pumice lapilli. 100 8.26 Strong brown and yellow brown med-fine pumice lapilli with scattered grey lithic lapilli. 150 Yellow brown fine ash with abundant yellow brown and grey lithic and pumice fine lapilli. 200 Grey, greyish brown and strong brown fine pumice and lithic lapilli, shower-bedded. 50 Yellow brown and brown fine ash with scattered pale brown and grey pumice and lithic fine lapilli. 150 8.76 94.21 , Waiohau Tephra Greyish brown and grey fine-med ash with scattered occasional yellow brown fine pumice lapilli. Pocketing white fine pumice ash within

A 55 centre of unit 10-20 mm thick. 10 Strong brown, soft fine pumice lapilli. 20 Grey coarse lithic ash. 20 Yellow brown fine ash with scattered strong brown fine pumice lapilli. 100 Strong brown fine, soft pumice lapilli. 50 Brown and grey fine ash with scattered common grey coarse lithic ash and strong brown fine pumice lapilli. 80 Pale yellow fine pumice lapilli with scattered grey fine lithic lapilli. 40 Pale grey and yellow fine ash with scattered fine yellow brown pumice lapilli. 100 9. 18 Pale yellow fine soft pumice lapilli with scattered grey lithic lapilli. 400 Grey and greyish brown bedded silts and sands with scattered and lenses of yellow brown fine pumice lapilli. 3000 T1 Eroded upper surface. Dark grey sandy matrix diamicton. Pebble large boulder clasts supported within matrix, clasts and matrix grey, massive and unbedded. 500-1000 T2, pumice sample -94.22 Sharp upper contact. Pale brown and pale yellow silt and sand matrix diamicton, with clasts of yellow, yellow brown and grey pumice and lithic pebble clasts supported within the matrix. 200 Grey and yellow pumice and lithic bedded med-fine sand, 20 mm scale. 1000 T2 Yellow, greasy silt matrix diamicton. Grey, and pinkish grey pumice and lithic pebble clasts supported within the matrix. Bright distinctive appearance. 1000+ T3 Firm reddish grey lithic sand matrix diamicton, with pebble-cobble clasts supported wi thin the matrix, massive and unbedded. Base obscured

Section 58 Makahikatoa Stream, T19 485221 True right bank of stream 500 m upstream of road. Description started at Mangamate Tephra.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

300 c. 4.0 Mangamate Tephra {MT), Grey and strong brown alternating beds of coarse-med poor1y Poutu Lapilli vesicular pumice and lithic ash, 30 mm shower-beds. 150 MT, Wharepu Tephra Strong brown , grey and greyish brown pumice and lithic fine-med lapilli, shower-bedded. 20 Strong brown fine pumice lapilli. 300 MT, Waihohonu Lapilli Grey and strong brown finepumice and lithic lapilli, shower-bedded, with grey dominated top and strong brown base. 20 Grey coarse-med lithic ash. 10 Strong brown coarse pumice ash. 10 Pale olive grey med ash. 60 4.57 MT, Oturere Lapilli Greyish brown , grey and strong brown fine-med pumice and lithic lapiili shower-bedded and ungraded. 40 Yellow brown and pale greyish brown fine-med ash. 30 Yellow brown and grey pumice and lithic coarse ash. 250-300 4.94 Pahoka Tephra Grey and olive brown fine-med pumice lapilli with scattered fine banded pumice lapilli. Shower-bedded but ungraded. 20 Brown fine-med ash with scattered grey and yellow brown fine lithic and pumice lapilli. 50 Yellow brown , pale brown and grey coarse pumiceand lithic ash in equal proportions. 10 Brown fineash with scattered yellow finepumice lapilli. 50 5.07 Bullet Formation, Pourahu Strong brown and yellow brown fine-med pumice lapilli, shower Member bedded with scattered grey lithic fine lapilli. 50 Greyish brown fine ash with scattered yellow brown and pale brown fine pumice lapilli. 50 5.17 Strong brown and yellow brown fine-med pumice lapilli with scattered grey lithic fine lapilli. 20 Greyish brown fine ash. 10 Grey coarse lithic ash. 10 Strong brown coarse pumice ash. 10 Greyish brown fine ash. 20 Strong brown fine pumice lapilli. 40 Greyish brown fine ash with scattered fine strong brown pumice iapilli. 10 Grey coarse lithic ash. 20 Strong brown and grey pumice and lithic finelapilli. 50 Brown and greyish brown fine ash. 100 5.46 Strong brown and yellow brown fine-med pumice iapiiliwith scattered grey fine lithic lapilli. 50 Yellow brown fine ash with scattered strong brown fine pumice lapilli. 20 Strong brown fine pumice lapilli.

A 56 100 5.61 Yellow brown and strong brown fine-med shower-bedded pumicelapilli with scattered grey lithic fine lapilli. 30 Brown fineash with scattered yellow brown fine pumice lapilli. 40 Greyish brown fine-med ash. 10 Strong brown fine pumice lapilli. 10 Grey coarse lithic ash. 10 Greyish brown fine ash. 10 Strong brown fine pumice lapilli. 10 Grey fine lithic lapilli. 50 Greyish brown fine ash. 20 Strong brown and yellow brown coarse pumice ash. 20 Yellow brown fine ash. 30 5.85 Yellow brown and grey fine pumice and lithic lapilli and coarse ash. 20 Strong brown fine pumice lapilli. 100 Grey and dark grey fine-med lithic ash with scattered strong brown and pale brown pumice fine lapilli. 200 Yellow brown and yellow med pumice lapilli with grey med-coarse lithic ash interbedded at the top of the unit. Distinctive unit in this area. 400 6.57 94.27 Waiohau Tephra Grey and pale greyish brown fine ash with scattered sparse yellow fine pumice lapilli. 10 mm cream-cakes of white fine pumice ash. 40 Yellow brown coarse pumice ash. 100 Greyish brown and yellow brown fine ash. 40 Strong brown and yellow brown fine pumice lapilli. 50 Yellow brown and strong brown fineash with scattered strong brown fine pumice lapilli. 500 Greyish brown bedded sands and silts, planar bedded on 10-30 mm scalewith interbedded yellow brown pumice pebbles in layers and lenses. 1000+ T1 Grey and strong brown sandy matrix diamicton. Clasts of sub-rounded and sub-angular pebble-large boulders supported within the matrix, grey clasts, unbedded. 8.30 Base into stream.

Section 59 Makahikatoa Stream, T19 495220

True right bank of stream c. 800 m downstream of State Highway 1 at crossing of power pylons. Description taken from distinctive grey topped pumice lapilli unit described near the base of the previous section (above Waiohau Tephra).

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

150 c. 8.0 Strong brown and yellow brown fine-med pumice lapilli with scattered grey lithic fine lapilli and grey fine-med ash mixed in at the top of the unit. 4000-5000 T1 Grey and greyish brown silty and sandy matrix diamicton, with clasts of pebble to boulder supported by the matrix. Clasts, grey and red lithic as well as abundant multi-coloured pumice and weathered lithic pebbles. 200-3000 T2, pumice sample - 94.31 Variable thickness, eroded top. Yellow silty matrix diamicton with grey lithic and white, pink and yellow brown pumice clasts supported by the firm silty matrix. Distinctive unit. 200 16.20 Grey mad-coarse lithic ash. 250 Pale greyish brown fine ash with scattered fine yellow and grey pumice and lithic lapilli. 20 Pale yellow and grey fine pumice and lithic lapilli. 40 Pale brownish grey fine ash with scattered fineyellow pumice lapilli. 50 16.56 Pale yellow and yellow brown fine pumice lapilli, soft with scattered grey fine lithic lapilli. 100 Pale grey fine ash, greasy. 30 16.69 94.28, Okareka Tephra Yellow brown and pale grey bedded silts, with white fine pocketing pumice ash. 400 Pale brownish grey bedded finesilts and sands with planar and wavy bedding on a 10-30 mm scale. 150 Grey fine ash with scattered fineyello w pumice lapilli. 20 Grey and pale yellow fine pumiceand lithiclapilli. 40 Pale grey fineash. 20 17.32 94.29, TeRere Tephra Pocketing wavy bedded fine white pumice ash. 250 Pale yellow and yellow brown finepumice lapilli with scattered weathered pale grey lithic lapilli. 40 Pale greyish brown fine ash with scattered pale grey finepumice lapilli. 50-100 Pale yellow and purplish grey finepumice and lithic weathered lapilli. 150 Pale greyish brown fine ash with scattered grey and pale yellow fine pumice and weathered lithic lapilli. 50 17.66 White and pale yellow brown finepumice lapilli with scattered grey lithic lapilli.

A 57 10 Greyish brown fine ash. 30 Pale yellow and grey fine pumice and lithic lapilli. 50 Greyish brown fine ash with scattered pale yellow fine pumice lapilli. 17.75 Base into Stream

Section 60 Makahikatoa Stream, T19 504215

True left bank of stream c. 2 km downstream of State Highway 1. 100 m downstream of telephone line crossing.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

c. 4.0 Upper partobscured 150 Strong brown fine pumice lapilli and coarse ash, with scattered grey lithic lapilli. 250 Greyish brown greasy fine ash. 40 Greyish brown med-coarse pumice ash. 50 Grey med-coarse lithic ash. 20 Yellow brown fine ash. 50 Grey and pale yellow brown coarse pumice and lithic coarse ash. 10 Brown fine ash. 50 4.63 Yellow brown coarse pumice ash and fine lapilli with scattered grey coarse lithic ash. 300 Hinuera Formation Greyish brown and brown fineash with scattered yellow and yellow brown fine pumice lapilli. Contains white pumice ash. 20 Yellow brown fine pumice lapilli. 10 Grey fineash. 40 Yellow brown fine pumice lapilli. 1200 6.20 Hinuera Formation Brown and pale brown bedded silt and fine pumice sand with scattered strong brown and pale brown fine pumice lapilli, also few layers and lenses of pumice lapilli. 400 Strong brown and grey sandy matrix diamicton, pebble-cobble rounded clasts supported within the matrix. 6000-8000 Kawakawa Tephra, Grey soft fine-med ash, massive, unbedded with very occasional Oruanui lgnimbrite cobble-pebble clast of andesite supported within. Scattered white Member coarse pumice ash. 10 Aokautere Ash Member Pinkish white fine ash. 10 Pale brown med pumice ash. 10 Pale yellow and white med-fine pumice ash. 20 14.65 Pinkish brown fine pumice ash. 2000- T4 Sharp upper contact. Brown and yellow brown silty matrix diamicton. 3000+ Pebble-boulder grey lithic clasts supported by the matrix. Some sites massive and unbedded others contain fluvial lenses and planar fabric. 17.65 Base into stream

Section Mangatawai64 Stream, T19 490239 True left bank of stream 50 m upstream of the State Highway 1 bridge over stream.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

Top obscured, coverbeds at road section. 200 T2 Brown silt and sand matrix diamicton, yellow strong brown, grey angular pebbly lithic clasts supported by the matrix. 300 Grey and greyish brown planar bedded sands and fine gravels, 30 mm scale beds. 1000 T3 Greyish brown sand and silt matrix diamicton. Pebble-cobble clasts supported by matrix, occasional boulders also present. Dominantly grey and few red lithic clasts with pumice pebble clasts also. Unbedded, massive. 10-20 Pale brown silt and fine sand. 1200 Grey, purplish grey, strong brown and brown fine-coarse sands and pebbles, pumice and lithic, planar bedded on 10-30 mm scale. 300 3.0 94 .41 , Rotoaira Tephra Strong brown coarse-med pumice lapilli with scattered lithic med lapilli. 600 Hinuera Formation Greyish brown, strong brown and grey fine-coarse bedded sands, planar bedded on a 10-30 mm scale. 1300 Grey and pale brown bedded sands and fine white pumice ash, planar and wavy bedded on a 10-30 mm scale. 1500+ T4 Greyish brown and grey sandy matrix diamicton, clast rich with cobble to boulder sized clasts supported within the matrix, grey lithic clasts. 6.40 Base into stream.

A 58 Section 66 Mangatawai Stream, T19 496238 True left bank of stream 600 m downstream from State Highway 1, at crossing of westem-most power pylon line.

Unit Depth Cum. Correlation and Samples Description (mm) depth (m) taken

200-1000 Taupo lgnimbrite White fine-coarse pumice ash with mixed pumice lapilli and blocks, local over-thickening, contains fragments of charcoal within base. 400 Mangatawai Tephra Black and purplish black 10-20 mm thick fine-med ash beds interbedded within dark brown, greasy, fine ash with paleosol development. 100 1.50 Papakai Formation (Pp) Brown and yellow brown greasy fine ash. 10-20 Waimahia Tephra within Pocketing white fine pumice ash within brown greasy fineash. Pp 400 Pp Brown and yellow brown fine ash with paleosol development, cracked vertically. 300-500 2.42 Mangamate Tephra (MT), Strong brown, and yellow brown fine-med poorly vesicular pumice and Poutu Lapilli lithic lapilli, shower-bedded and ungraded. 4000 T1 Grey and greyish brown silty and sandy matrix diamicton. Pebble large boulder clasts supported by matrix, grey sub-angular lithic clasts, massive unbedded. 300 Unconformable angular upper eroded contact. Pale purplish brown, pale brown and pale yellow bedded silt and fine sand, wavy and planar bedded on a 10 mm scale,with scattered fine pale yellow pumice lapilli. 150 6.87 Brown and dark brown fibrous lignite, with woodfrag ments, wavy and planar laminations on a 5-20 mm scale. Mgt(a) taken 20 mm from top, and (b) taken 20 mm above base. 200 Pale grey bedded silt and fine sand, planar and wavy bedded on a 10- 20 mm scale, with scattered pale yellow fine pumice lapilli present. 50 7.12 94.43 Pale yellow fine, soft, pumice lapilli with scattered sparse grey lithic fine lapilli. 100 Pale yellow and pale grey silt and finesand, planar and wavy bedded on a 10-20 mm scale, with scattered pale yellow fine pumice lapilli present. 30 7.25 94.44 Pale yellow fine pumice lapilli with few scattered pale grey lithic fine lapilli. 50 Pale yellow and pale grey silt and finesand, planar and wavy bedded on a 10-20 mm scale. 80 7.38 94.45 Yellow and pale yellow fine pumice lapilli with few scattered pale grey lithic fine lapilli. 10 Purplish brown firm lignite. 50 7.44 94.46 Pale brown fine-med pumice ash. 300 Dark brown and brown firm lignitewith large (>500 mm long, >200 mm diameter) woodfr agments within. Wood sampled as (c) and lignite sampled in 50 mm intervals as (d)-(h). 50 Grey bedded fine sand planar 10 mm scale bedding. 100 Dark brown and brown lignite planar bedded, on a 10-30mm scale, firm. (I) sampled near top and 0) near base. 50 7.94 94.47, Rerewhakaaitu White fine -med pumice ash. Tephra 150 Black and dark brow lignite, firm, bedded, planar and wavy bedding on a 10-30 mm scale. Sampled 2 cm from top and 2 cm from base. 10 Pale brown wavy bedded fine silt. 400 8.50 Dark brown and brown lignite and organic rich silts and fine sands. Sampled in the zones of most pure lignite. 1200 Pale grey and pale yellow bedded silts and fine sands with scattered pale yellow fine pumice lapilli. Planar and wavy bedded on a 5-30 mm scale, very finely laminated at base. 200-400 T3 Pale brown sandy matrix diamicton, pebble-cobble grey lithic clasts supported by matrix, unbedded. 1200 Massive, unbedded pale brown and pale olive silt and fine sand. 1500+ T3-T4 Greyish brown and brown silt and sandy matrix diamicton, clasts of grey and greyish brown lithic pebble-boulders supported by matrix, unbedded. 12.80 Base into stream.

A 59 Section 68 Mangamate Stream, T19 505248

True left bank of stream 150-200 m upstream of crossing of eastern-most line of power pylons, c. 1 km downstream of State Highway 1.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

400 Mangatawai Tephra Dark brown fine greasy ash with interbedded purplish grey and black fine-med ash beds 10-20 mm thick. 150 Papakai Formation (Pp) Brown and yellow brown fine greasy ash. 50 0.60 Hinemaiaia Tephra within Scattered white pumice med-coarse ash within brown finegreasy ash. Pp 250 Pp Brown and yellow brown fine ash wi th scattered grey and strong brown pumice and lithic lapilli in base of unit. 500 1.35 Mangamate Tephra (MT). Strong brown and grey fine-med pumice and lithic lapilli, shower- Poutu Lapilli bedded but ungraded. 150 MT, Waihohonu Lapilli Grey and strong brown med-coarse poorly vesicular pumice and lithic ash. 250 Strong brown and yellow brown and grey pumice and lithic lapilli. 50 Grey and greyish brown fine-med ash, 20 mm onto 30 mm of strong brown fine-med ash. 350 2.15 MT, Oturere Lapilli Strong brown, greyish brown and grey fine-med poorlyvesicular pumice and lithic lapilli, shower-bedded but ungraded. 20 Grey fine lithic lapilli. 50 Yellow brown fine ash. 30 Strong brown fine pumice lapilli and coarse ash. 300 2.55 Pahoka Tephra Upper 150 mm grey and greyish brown coarse pumice ash, shower- bedded, lower 150 mm olive grey and pale brown fine-med pumice lapilli, with scattered banded pumice lapilli. 300 Greyish brown fineash with scattered strong brown and grey fine pumice and lithic lapilli. 20-50 2.90 Yellow and yellow brown fine pumice lapilli, with scattered grey lithic lapilli. 50 Greyish brown and brown fine-med ash. 20 Strong brown and grey fine pumice and lithic lapilli. 50 Brown and grey fine-med ash. 10 Strong brown pumice lapilli. 50 Yellow brown and brown fine ash. 80 3. 18 Strong brown fine pumice lapilli, wi th scattered grey fine lithic lapilli. 50 Strong brown fine ash with scattered fine pumice lapilli. 250 Grey and strong brown fine pumice and lithic lapilli, shower-bedded and ungraded. 20 3.50 94.58, Waiohau Tephra Greyish brown fine-med ash with scattered grey and strong brown fine pumice and lithic lapilli, 30-40 mm cream-cakes of white finerhyolitic ash interbedded. 300 Strong brown, greyish brown and grey med-coarse pumice and lithic ash, shower-bedded. 40 Strong brown fine ash. 100 3.94 Strong brown and reddish brown fine pumice lapilli with scattered fine grey lithic lapilli. 30 Strong brown and brown fine ash. 20 Strong brown fine pumice lapilli. 200 4.19 Rerewhakaaitu Tephra Brown and greyish brown fine ash with interbedded cream-cakes of white fine pumice ash. 150 Yellow brown and strong brown normally graded med ash-fine pumice lapilli, with scattered grey lithic fine lapilli and coarse ash. 50 Grey, greyish brown and strong brown fine-med pumice ash. 150 Strong brown fine pumice lapilli. 200 4.74 Hinuera Formation Brown, pale brown and white reworked fine ash with scattered fine strong brown pumice lapilli. 300 Grey med lithic ash with scattered strong brown fine-med pumice lapilli. 200 Pale grey soft greasy fine ash with scattered yellow brown and strong brown fine pumice lapilli. 500 5.74 Yellow brown and strong brown finepum ice lapilli, reworked, with scattered grey fine lithic lapilli. 200 Pink and pale yellow fineash with scattered strong brown fine pumice lapilli. 1000 T4 Grey sandy matrix diamicton, pebble to boulder clasts supported within the matrix, massive and unbedded. 3000+ Sharp upper contact. Pale greyish brown and brown silty matrix diamicton. Sparse pebbly multi-coloured lithic and pumice clasts supported by matrix, massive and unbedded. 9.94 Base into stream.

A 60 Section 72 Tongariro River, T19 356556 True left bank of river immediately downstream of "Sand Pool".

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

400 Grey bedded sand with scattered fine white pumice lapilll pebbles. Recent soil developing into top of unit, greyish brown and brown fine coarse sand. 400 0.80 Taupe lgnlmbrite Fine-coarse pumiceash with mixed pumice lapilli and blocks, unconformable sharp lower contact. 2500-3000 3.80 Grey and pale brown diamicton, sand and finepebble matrix supporting grey clasts of cobble to boulder size. Sub-rounded clasts, with a more well sorted matrix than unit below. Occasional interbedded lenses of sand and fine cross-bedded gravels. 4000+ Greyish brown and grey sandy matrix diamicton. Surface reworked in many places by river and finesdepleted. Pebble-huge sub-rounded and sub-angular boulder clasts supported by matrix. Fine clasts grey, yellow and red lithic and occasional pumice larger clasts all grey lithic. Occasional huge boulder clasts up to 3 m diameter. Very firm, cemented? matrix. 7.80 Base into stream.

Section 73 Tongariro River, T19 364544 True left bank of river, large cliff section at "Breakaway Pool"

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

5000 Taupe lgnimbrite White pumice fine-coarse ash with mixed pumice lapilli and large pumice blocks, charcoal fragments and logs near base of unit. Upper 800 mm black present soil developed. Unconformable lower contact. 300 5.30 Mangamate Tephra, Poutu Strong brown and yellow brown fine pumice lapilli, with scattered grey Lapilli lithic lapilli, shower-bedded and ungraded, irregular upper contact. 600-800 94 .79, Karapiti Tephra Yellow brown and pale yellow brown fine ash with paleosol development and paleo-root channels, white fine ash pocketing 10 mm near top. 800-1 000 7.10 Grey sandy matrix diamicton, pebble-cobble clasts supported by matrix, cross bedded In places and with only faint planar fabric in other places. 800-1 200 Grey and olive grey massive coarse-fine sand, poorly sorted, weak planar fabric, possible log casts present. 20 Wavy bedded pale brownish grey silt and fine sand. 1500 9.82 Grey, olive grey and pale greyish brown coarse-fine poorly sorted matrix dlamlcton, planar fabric with occasional cobble and small boulder clasts supported by matrix,emp ty log castsseen. Lower 500 mm massive with no planar fabric. 100-150 Pale greyish brown silt, massive with paleo-root channels. 1000 Grey and brownish grey fine-coarse sand, massive with faint planar fabric In places. 2000 12.97 Grey and greyish brown fine-coarse sand matrix dlamicton with pebbly clasts supported by matrix, grey red and yellow brown pebble clasts. Planar fabric, and poorlysorted. 150 Pale grey fine sand, planar bedded. 30 Dark grey fine sand and silt with scattered yellow brown fine pumice lapilli. 30 Grey and pale brown planar bedded sand and silt. 50 Rotoaira Tephra Strong brown and pale brown fine pumice lapllli with scattered grey lithic fine lapllli. 10 Dark grey fine-med lithic ash. 40 13.28 Grey, strong brown and yellow brown finepumic e and lithic lapilli, grey at top and strong brown and yellow brown soft pumice lapllli at base. 200 Hinuera Formation Greyish brown and grey fine sand and silt with mixed in fine white pumice ash. 30 Grey and greyish brown finesand and silt. 50 Pale brown fineash with scattered grey and yellow brown finepum ice and lithic lapllli. 300 Greyish brown and grey sand and silt, planar bedded. 0-1 700 14.56 Kawakawa Tephra/ Pinkish pale brown fine ash with scattered coarse and med pumice Oruanui lgnimbrite ash within, occasional pumice clasts up to fine lapllll In size. Member 20 Aokautere Ash Member Coarse white pumice ash

A 61 30 Pale pinkish brown fine ash with accretionary lapilli up to 10 mm in diameter. 40 White fine pumice lapilli, and accretionary lapilli within white fine pumice ash. 10 Pale brown fine pumice ash. 20 White fine pumice lapilli. 200-250 White and pale brown fine pumice ash with abundant accretionary lapilli 5-20 mm in diameter. 10 Brown fine ash. 4000+ Greyish brown sandy and silty matrix diamicton. Clasts of sub-angular pebble-boulders supported within the firmly cemented matrix. 18.94 Base into river.

Section 76 Tongariro River, T19 533383 Road cutting on the western side of Sate Highway 1, 200 m south of the Trout Hatchery.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

300-400 Taupo lgnimbrite White and pale grey fine-coarse pumice ash with mixed white pumice lapilli and blocks, with present soil development. 400-450 0.85 Mangamate Tephra, Poutu Strong brown and yellow brown fine pumice lapilli, shower-bedded and Lapilli ungraded with scattered grey fine lithic lapilli. 30 Pale grey med-coarse lithic ash. 100 Yellow brown and greyish brown coarse pumice ash and finelapilli. 150 Greyish brown and yellow brown fine-med ash. 100 1.23 Yellow brown and strong brown finepum ice lapilli and coarse ash, with scattered grey lithic lapilli. Grey lithic fine lapilli rich zone near top of unit, 10 mm thick. 100 Pale brown fine ash. 10 Pale brown fine pumice lapilli. 450 Pale brown and yellow brown fine ash with paleosol development and scattered grey and yellow brown fine pumice and lithic lapilli. 40 1.83 Strong brown and yellow brown fine pumice lapiln 10 Dark grey med-fine ash. 10 Pale brown fine ash. 10 Dark grey med-fine ash. 40 Rotoaira Tephra Strong brown fine pumice lapilli and coarse ash with scattered grey fine lithic lapilli. 10 Grey fine ash. 30 1.94 Strong brown and yellow brown finepumice lapilli and coarse ash. 50 Pale brown fine ash. 2000+ Hinuera Formation White, grey and pale pinkish brown pumice rich sands silts and gravels, cross-bedded and planar bedded on a 20-50 mm scale. 3.99 Base obscured.

Section 78 Tongariro River, T19 404533 Disused pumice quarry at end of Admirals Reserve Road, off State Highway 1, 2 km south of Turangi.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

1000-6000 Taupo lgnimbrite White and pink fine-coarse pumice ash with mixed pumice lapilli and blocks and fragments of charcoal near base. Variable thickness and eroded irregular lower contact. 100 6.1 Mangamate Tephra, Poutu Strong brown, yellow brown and grey fine pumice and lithic lapilli, Member shower-bedded and ungraded, with scattered grey fine lithic lapilli. 100 Grey and greyish brown coarse pumice and lithic ash and fine lapilli. 1200 Brown and yellow brown fineash with scattered yellow brown fine pumice lapilli, paleosol development and paleo-root channels present. 50 Grey and strong brown fine pumice and lithic lapilli and coarse ash. 100 Pale brown fine ash. 500-2000+ Kawakawa Tephra/ Pinkish brown and pale brown fine-med pumice ash with scattered Oruanui lgnimbrite white coarse pumice ash, massive, unwelded but firm. Member 9.55 Base obscured.

A 62 Section 81 Oturere Stream, T19 430238 Exposure on the top of a lava flow on the true right side of Oturere valley, 500 m due west of Oturere Hut, within wilderness area.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

100 Brown fine ash with scattered grey lithic coarse ash. 400 Mangamate Tephra, Grey and greyish brown fine lithic and poorly vesicular pumice lapilli Oturere Lapilli and coarse ash, shower-bedded and ungraded. 300 0.80 Pahoka Tephra Pale grey and pale olive brown fine-med pumice lapilli, shower- bedded, ungraded, with distinctive common banded pumice lapilli. 20 Dark grey fine lithic lapilli. 20 Brown fine ash. 10 Grey med-fine ash. 150 Yellow and grey fine pumice and lithic lapilli and coarse ash, reversely graded with equal proportions of pumice and lithics. 200 1.20 Bullet Formation, Pourahu Greyish brown fine pumice lapilli, shower-bedded with scattered grey Member fine lithic lapilli. 50 Brown fine ash, with scattered pale yellow brown and grey fine pumice and lithic lapilli. 150 Greyish brown fine pumice lapilli, shower-bedded, with scattered fine grey lithic lapilli. 50 Grey and purplish grey fine lithic lapilli and coarse ash. 150 Pale greyish brown fine-med pumice lapilli, with scattered grey fine lithic lapilli, 10 mm grey med lithic ash zone 50 mm from top of unit. 20 Grey and pale brown fine-med ash. 300 1.92 Greyish brown and grey pumice and lithic coarse ash and fine lapilli. 50 Grey and yellow brown banded 10 mm layers of alternating colour, pumice and lithic coarse ash. 10 Yellow brown coarse pumice ash. 10 Dark grey fine-med ash. 10 Yellow coarse pumice ash. 20 Grey lithic and pumice coarse ash. 150 2.17 Yellow brown coarse pumice ash wi th scattered coarselithic ash, shower-bedded. 150 Dark grey and greyish brown alternating zones of med ash and fine pumice and lithic lapilli, shower-bedded. 20 Yellow brown coarse pumice ash. 10 Grey coarse lithic ash. 20 Yellow brown coarse pumice ash. 10 Dark grey coarse lithic ash. 30-50 2.43 95.2, Waiohau Tephra Grey med-coarse lithic ash with pocketing white fine pumice ash cream-cakes. 10 Yellow brown coarse pumice ash. 10 Dark grey coarse-med lithic ash. 10 Yellow brown coarse pumice ash. 10 Purplish grey med ash. 100 Yellow brown and greyish brown zones of shower-bedded fine pumice lapilli. 200 Greyish brown and grey alternating zones of fine-med pumice lapilli, shower-bedded. 10 Greyish brown fine ash. 1000 3.78 Yellow brown, yellow and pale greyish brown fine pumice lapilli, shower-bedded with scattered grey lithic fine lapilli. 50 Grey and greyish brown coarse pumice and lithic ash. 30 Greyish brown fine pumice lapilli. 300 4.16 95.3, Rerewhakaaitu Greyish brown and grey shower-bedded coarseash and fine pumice Tephra lapilli, alternating coloured zones, 5 mm cream-cakesof white fine pumice ash 30 mm from base. 10 Pale grey fineash. 200 Greyish brown and grey coarse pumice and lithic ash and fine pumice lapilli, shower-bedded. 10 Greyish brown fine ash. 500 Greyish brown coarse pumiceash and fine lapilli with scattered grey fine lithic lapilli. 20 Yellow brown fine pumice lapilli with scattered grey fine lithiclapilli. 200 5.10 Grey and greyish brown med-coarse pumice and lithic ash, with scattered yellow brown fine pumice lapilli. 10 Grey fine ash. 30 Greyish brown coarse pumice ash. 10 Brown fine ash. 30 Greyish brown coarse pumice ash. 20 Greyish brown and yellow brown fine pumice lapilli.

A 63 10 Greyish brown coarse pumice and lithic ash. 10 Yellow brown coarse pumice ash 10 Dark grey fine ash. 70 5.30 Greyish brown coarse pumice ash and fine lapilli. 10 Brown and yellow brown fine ash. 40 Greyish brown and yellow brown med ash and finepumic e lapilli intermixed. 300 5.65 20-30 mm of dark grey fine-med ash intermixed within the top of yellow brown and yellow coarse-fine pumice lapilli, shower-bedded and reversely graded, with scattered grey finelithic lapilli. 10 Grey fine ash. 100 Yellow brown and strong brown fine pumice lapilli. 150 Greyish brown and grey med-coarse sand with interbedded pale brown pumice and grey pebbles and cobbles. 200 6.11 Strong brown and yellow brown coarse-med pumice lapilli, with scattered grey fine lithic lapilli. 100 Olive greyish brown fine pumice lapilli, with scattered grey lithic fine lapilli. 150 Grey and greyish brown med-coarse sands with interbedded greyish brown and grey lithic and pumice pebbles. 4000+ 10.36 Grey lava flow, base not seen.

Section 82 Mangatetipua Stream, T19 423339 Road side section on the southern side of State Highway 4 7 A, large lava flow exposure 500 m west of road bridge over Mangatetipua Stream.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

200-3000 Taupo lgnimbrite White fine-med pumice ash mixed with white pumice lapilli and blocks, primary in some places, bedded and reworked in other places. 500 3.50 Mangatawai Tephra Brown and greyish brown fine ash, purplish grey and black fineash within lower 100 mm. 100 Papakai Formation (Pp) Brown and yellow brown fine ash, greasy with paleosol development. 10-20 3.62 Waimahia Tephra within White fine pumice ash cream-cakes within brown fine ash. Pp 300 Pp Yellow brown fine ash. 50 3.97 Hinemaiaia Tephra within White coarse-med pumice lapilli scattered within finebrown ash. Pp 300 Pp Yellow brown fine ash, greasy. 500-800 5.07 Mangamate Tephra (MT). Strong brown and yellow brown fine-med pumice lapilli, shower- Poutu Lapilli bedded and ungraded, with scattered grey fine lithic lapilli, upper part mixed within lower part of Papakai Tephra. 30 Grey med-coarse lithic ash. 50-100 Brown fine ash with scattered grey and yellow brown finelithic and pumice lapilli. 400 MT, Te Rato Lapilli Grey and greyish brown poorly vesicular fine-med pumice lapilli, shower-bedded and ungraded. 0-2000 7.60 95.6 Waiohau Tephra Variable thickness yellow brown and brown fine ash, with scattered yellow brown pumice lapilli and occasional large lava ciast, colluvial layer. Unconformable upper contact. Two apparent horizons of white fine pumice ash cream-cakes. 7000+ 95.8 Grey and reddish grey scoriaceous and dense lava flow. 14.60 Base obscured.

Section 83 Tahurangi/ Te Karo Streams, T19 448326 Large road cutting on the southern side of State Highway 47A, 500 m east of Rotoaira Trust Fishing campjetty.

Unit Depth Cum. Formation and Member Description (mm) depth (m)

200-300 Taupo lgnimbrite White fine-coarse pumice with mixed pumice lapilli. 300 Mangatawai Tephra Brown fine ash upper 150 mm, lower 150 mm grey and dark grey med- fine ash beds 10-30 mm thick interbedded with brown fine ash .. 200 Papakai Formation (Pp) Brown and yellow brown fine greasy ash. 50 0.85 Hinemaiaia Tephra within White mad-coarse pumice ash scattered within yellow brown fine ash. Pp 50-300 Pp Yellow brown and brown fine greasy ash. 400-700 Mangamate Tephra (MT), Variable upper contact, yellow brown , strong brown and grey fine Poutu Lapilli pumice and lithic lapilli, shower-bedded and ungraded. 30 Grey coarse-mad ash. 20 Greyish brown fine ash. 400 2.30 MT, Te Rato Lapilli Grey and pale greyish brown fine pumice and lithic lapilli, shower-

A 64 bedded and ungraded. 100 Pale brown and yellow brown fineash with scattered yellow brown and strong brown fine pumice lapilli. 200 Yellow brown fine ash. 20 Strong brown fine pumice lapilli. 20 Grey med-fine ash. 20 Strong brown fine pumice lapilli and coarse ash. 150 Greyish brown and brown fine ash with scattered strong brown fine pumice lapilli and coarse ash. 10 2.78 95.9, Waiohau Tephra Pocketing white fine pumice ash cream-cakes, within greyish brown fine ash. 10 Black med lithic ash, pocketing within greyish brown fine ash. 20 Brown fine ash wi th scattered yellow brown pumice and grey lithic coarse ash. 40 Strong brown and yellow brown coarse pumice ash. 50 Greyish brown and yellow brown fine ash with scattered yellow brown fine pumice lapilli. 100 3.00 Yellow brown and strong brown fine pumice lapilli with brown fine ash intermixed toward base. 150 Greyish brown fine-med ash. 200-250 Rotoaira Tephra Yellow brown and strong brown fine-med, soft pumice lapilli, with scattered grey fine lithic lapilli. 30 Black and dark grey med lithic ash. 80 3.51 Yellow brown and strong brown fine-med pumice lapilli, with scattered fine grey lithic lapilli. 20 Greyish brown med ash. 100 Yellow brown sand with interbedded grey and strong brown pebbles. 100 3.73 Kawakawa Tephra, Pinkish pale brown fine-med pumice ash. Aokautere Ash Member 10 White med pumice ash. 10 Pale pink and pale grey fine pumice ash. 40 Alternating layers of pale yellow and pale pink fine and med pumice ash, with basal 10 mm fine pale pink ash. 400 Grey and greyish brown med sand, massive and well sorted. 20 Strong brown soft pumice lapilli. 400 Dark grey fine-coarse bedded sands, planar bedded. 1000+ 4.61 95.10 Grey lava flow, base obscured.

Section 88 Unnamed stream between Tahurangi and Rahuituki Streams, T19 475304 True left bank of stream 20 m upstream of forestry road bridge, Te Ngoi forestry road.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

400 Mangamate Tephra (MT), Strong brown fine pumice lapilli with scattered grey lithic finelapilli, Poutu Lapilli shower-bedded and ungraded. 70 Grey med lithic ash. 30 Greyish brown fine ash with scattered yellow brown fine pumice lapilli. 20 Grey med-fine ash. 10 Yellow brown coarse pumice ash. 60 0.59 MT. Oturere Lapilli Grey med lithic ash, grading into 30 mm yellow brown coarse pumice ash with scattered grey coarse lithic ash. 30 Greyish brown fine ash with scattered yellow brown fine pumice lapilli. 100 0.62 MT. Te Rato Lapilli Greyish brown and yellow brown fine pumice lapilli. 80 Greyish brown and pale brown fine ash with scattered finegrey and yellow brown pumice and lithic fine lapilli. 150 Yellow brown and strong brown coarse pumice ash. 250 Greyish brown fineash with scattered yellow brown fine pumice lapilli. 100 1.10 Strong brown fine pumice lapilli, with scattered grey fine lithic lapilli, 10 mm grey lithic dominated layer near top of unit. 30 Greyish brown fine ash with scattered yellow brown finepumice lapilli. 50 Yellow brown fine pumice lapilli. 600 Greyish brown fine ash with scattered yellow brown finepumic e lapilli. 100 1.88 Yellow brown and strong brown fine pumice lapilli, soft, with scattered fine grey lithic lapilli. 400 Greyish brown fine ash with scattered yellow brown finepumice lapilli. 40 Grey med lithic ash. 100 2.42 Strong brown fine pumice lapilli with scattered grey ash. 20 Grey med ash. 200 Yellow brown fine-med pumice lapilli with scattered grey fine lithic lapilli. 10 Grey med lithic ash. 100 2.75 Yellow brown and strong brown fine pumice lapilli with scattered grey fine lithic lapilli.

A 65 100 Grey med lithic ash. 100-300 Kawakawa Tephra, Pale brown fine-med massive unbedded ash with scattered coarse Oruanui lgnimbrite pale brown and white pumice ash. Member 500+ T4 Boulder rich diamicton, grey sandy matrix, firm and cemented with pebble-boulder clasts supported within, large grey boulder clasts. 3.65 Base into stream.

Section 89 Poutu Canal, T19 497327 True left side of canal as it flows into Poutu Dam (basal part exposed 300 m further upstream on true right side of canal).

Unit Depth Cum. Formation and Member Description (mm) depth (m)

Top Obscured. 200 c. 2.0 Brown fine ash with scattered strong brown and yellow brown fine pumice lapilli. 30 Rotoaira Tephra Yellow brown and strong brown fine pumice lapilli. 10 " Grey med-coarse lithic ash. 120 Strong brown and yellow brown fine pumice lapilli, with scattered grey fine lithic lapilli. 200 Greyish brown fine ash with scattered strong brown fine pumice lapilli. 40 Strong brown and yellow brown fine pumice lapilli. 10 Grey coarse lithic ash. 50 2.46 Yellow brown and strong brown fine pumice lapilli with scattered grey fine lithic lapilli, shower-bedded. 10 Grey lithic fine-med ash. 30 Strong brown and yellow brown fine pumice lapilli. 250 Greyish brown with scattered grey and yellow brown finepumice and lithic lapilli. 1000 Hinuera Formation Grey and pale brown lithic and pumice sands and fine gravels, rich in pale brown pumice ash. Grey and brown pumice and lithic gravels, fine-coarse bedded in planar layers and lenses. 500 4.25 Kawakawa Tephra/ Pale brown and white fine pumice ash with scattered white fine pumice Oruanui lgnimbrite lapilli, massive and unbedded, uneven upper contact. Member 15000+ T4 Grey sandy matrix diamictons, pebble-boulder grey lithic clasts supported by matrix. Many units, some massive and unbedded others with weak planar fabric. 4.75 Base obscured.

Section 91 Puketarata Stream, T19 528289 True left bank of stream, below State Highway 1 bridge.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

Top obscured. 600 c. 4.0 Brown and greyish brown fine ash with scattered yellow brown fine pumice lapilli. 20 Yellow brown and strong brown fine pumice lapilli. 200 Greyish brown fine greasy ash. 50 Grey and greyish brown fine-med lithic ash. 30 Pale brown and greyish brown fineash. 50 4.35 Rotoaira Tephra Strong brown and yellow brown fine pumice lapilli with scattered fine lithic lapilli. 20 Grey fine ash. 50 Strong brown fine pumice lapilli with scattered grey fine lithic lapilli. 10-20 Grey fine-med ash. 50 Yellow brown and strong brown fine pumice lapilli, with scattered grey fine lithic lapilli. 250 Greyish brown and yellow brown fine, greasy ash with paleosol development. 1000+ T3/T4 Grey and brownish grey sandy matrix diamicton with pebble to large boulder clasts supported by matrix. Upper part reworked and fines depleted, clast supported with bedded grey and strong brown sand between grey lithic clasts. 5.74 Base into stream.

A 66 Section 93 Mangahouhounui Stream, T19 51231 5 True right bank of stream where it is crossed by the Poutu canal, deep partially obscured section.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

500 Mangamate Tephra, Poutu Strong brown and yellow brown fine pumice lapilli, shower-bedded Lapilli and ungraded, scattered grey fine lithic lapilli. 2000 Partially obscured ash and lapilli beds, undescribed 2000 Hinuera Formation Brownish grey, pale brown and grey bedded sands and gravels, planar and cross-bedded containing abundant pale brown and white pumice ash. 10000+ T4 Greyish brown and dark grey sandy matrix diamicton, massive and unbedded, pebble-large boulder clasts, grey and reddish grey angular lava clasts supported by the matrix. Potentially several diamicton units present. 14.50 Base into stream.

Section 95 Tongariro River, T19 540397 Stag Pool, true right bank of river.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

4000 Taupo lgnimbrite White fine-coarse pumice ash with mixed pumice lapilli and blocks, Charcoal fragments and logs in base of unit. 500-2000 Variable lower contact pinches over paleo-topography planar upper contact. Grey and greyish brown sandy matrix diamicton with pebble boulder clasts supported within the matrix. 0-1500 Lenses of bedded grey and greyish brown gravels and sands, rounded grey gravels and grey and yellow brown sands. 4000 Large boulder rich diamictons. Grey and greyish brown firm sandy matrix supporting clasts of pebble-boulder, grey sub-rounded boulders, two or more units. 11.50 Base into river.

Section 96 Tongariro River, T19 549328 Road cutting along Rangipo Prison Farm road toward river, northern side of road, 300 m before bridge over river.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

300-1000 Taupo lgnimbrite White fine-coarse ash with mixed pumice lapilli and blocks, charcoal fragments within base. 200 Mangatawai Tephra Dark brown fine ash with interbedded grey and purplish grey fine-med ash, pocketing in the base of the unit. 200 Papakai Formation Brown fine greasy ash with paleosol development, scattered white coarse pumice ash. 200-400 1.80 Hinemaiaia Tephra White and pale brown coarse pumiceash shower-bedded occasionally over-thickened. 500 Papakai Formation Brown and yellow brown fine ash with grey fine pumice and lithic lapilli scattered within base of unit. 200-400 Mangamate Tephra (MT), Variable upper contact, strong brown and yellow brown finepumice Poutu Lapilli lapilli with scattered grey lithic lapilli, shower-bedded but ungraded. 10 Grey med lithic ash. 30 Greyish brown fine ash. 10-20 2.76 Poronui Tephra White pocketing fine pumice ash, within grey finegreasy ash. 50 Greyish brown fine ash. 20 Grey med lithic ash. 20 Greyish brown fine ash. 10 Strong brown, yellow brown and grey finepum ice and lithic lapilli. 30 2.89 MT, Te Rato Lapilli Grey coarse ash and finepumice and lithic lapilli. 40 Greyish brown fine ash with scattered grey finelithic lapilli. 20 2.95 Pahoka Tephra Grey and olive pale brown fine pumice lapilli, with scattered occasional banded lapilli. 100 Greyish brown and brown fine, greasy ash with scattered yellow brown fine pumice lapilli. 500 Planar bedded gravels and sands, poorly sorted coarse and finegrey gravels and sand, sub-rounded and sub-angular, with scattered yellow brown fine pumice pebbles. 40 Greyish brown fine-med sand. 10-200 Sandy matrix pebbly diamicton, lenzoidal. Grey and greyish brown

A 67 pebbles wi thin greyish brown sandy matrix. 50 Greyish brown fine-med sand. 400-700 4.54 Grey and greyish brown planar bedded sands and gravels, 20-30 mm scale beds, poorly sorted, pebbles rich in layers and lenses. 20 Greyish brown fine ash. 20 Greyish brown and yellow brown fine pumice lapilli. 40 Greyish brown fine ash. 600 Grey and greyish brown sandy matrix diamicton, pebble-cobble clasts supported within the matrix, clast rich unit, massive and unbedded. 50 Brown fine ash. 40 5.31 Yellow brown and strong brown fine pumice lapilli with scattered grey fine lithic lapilli. 30 Brown fine ash. 60 Grey and strong brown fine lithic and pumice lapilli. 20 Greyish brown fineash. 10 5.43 95. 13 Waiohau Tephra White fine pumice ash, pocketing in places and a continuous layer in other places. 250 Brown and gleyed pale brown fine ash with scattered pale brown coarse pumice ash. 30 Strong brown and grey coarse-med pumice and lithic ash. 200 5.91 Brown fine ash with paleosol development and root channels present. 15000+ Base of section seen at road bridge across the river. Greyish brown and grey med-coarse sand-matrix diamicton, pebble-large boulder clasts, massive and unbedded, sparse large boulders, firm unit. 20.91 Base into stream.

Section 97 Tongariro River/ Whitikau Stream, T19 558336 Road cutting on the western side of the Rangipo Prison farm road next to the bridge across the Whitikau Stream.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

400 Taupo lgnimbrite White and pale grey fine-coarse pumice ash with mixed pumice lapilli and blocks. 50 Mangatawai Tephra Dark brown and grey fine-med ash. 200 Papakai Formation (Pp) Brown and yellow brown fine ash 200 0.85 Hinemaiaia Tephra Pale yellow brown med-coarse pumice ash, shower-bedded. 300 Pp Brown fine greasy ash with occasional white med pumice lapilli. 20 1.17 Motutere Tephra Pale brown fine pumice ash, pocketing within brown fine ash. 150 Pp Brown fine greasy ash with scattered grey fine lithic and pumice lapilli. 400-500 Mangamate Tephra (MT), Strong brown and fine pumice lapilli, shower-bedded and ungraded Poutu Lapilli with scattered grey fine lithic lapilli. 100 1.92 Poronui Tephra Greyish brown and grey fine-med ash with pocketing 10 mm thick fine white pumice ash. 30 Grey med lithic ash. 20 Greyish brown fine ash. 40 MT Grey and yellow brown coarse pumice ash and fine lapilli. 100 2.11 Pahoka Tephra Grey and pale olive brown fine pumice lapilli and coarse ash. 250 Greyish brown fine ash with scattered fineyello w brown fine pumice lapilli. 300-500 Grey and greyish brown sandy matrix diamicton, pebble sub-angular clasts supported within the matrix, planar fabric preserved. 20 Greyish brown fine ash with scattered fineyellow brown fine pumice lapilli. 40 Strong brown and yellow brown fine pumice lapilli. 20 Greyish brown fine ash with scattered fine yellow brown fine pumice lapilli. 40 2.98 Grey and strong brown coarse pumice and lithic ash and fine lapilli. 20 Greyish brown fine ash with scattered fine yellow brown fine pumice lapilli. 10 3.01 Waiohau Tephra White fine pocketing fine pumice ash, within greyish brown fine ash. 250 Greyish brown fine ash with scattered fineyellow brown fine pumice lapilli. 40 Strong brown and grey coarse pumiceand lithic ash and fine lapilli. 150 Greyish brown fine ash with scattered fineyello w brown fine pumice lapilli. 10-20 Yellow brown and grey fine pumice lapilli. 150 3.62 Greyish brown fine ash with scattered fine yellow brown fine pumice lapilli. 20 Strong brown coarse pumice ash. 30 Grey and strong brown coarse lithic and pumice ash and strong brown fine pumice lapilli. 20 Grey med lithic ash. 80 3.77 Strong brown and yellow brown fine pumice lapilli with scattered grey

A 68 fine lithic lapilli. 20 Dark brown greasy fineash. 100 Pale brown gleyed fine greasy ash. 600 Brown greasy fine ash. 500+ 3.99 Kawakawa Tephra, Pale brown and pink fine pumice ash with scattered white coarse Oruanui lgnimbrite pumice ash Member 10000- 95. 16 Grey andesite lava flow. 15000 18.99 Base into stream.

Section 98 Tongariro River, T19 53741 1 True right bank of river between "The Rip" and "Hydro Pool" beside the walkway.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

400 Taupo lgnimbrite White fine-coarse pumice ash with mixed white pumice lapilli and blocks. 250 Grey and greyish brown sands and gravels. 300 Grey and greyish brown med sands, planar and wavy bedded. 50 Papakai Formation (Pp) Yellow brown finegreasy ash with paleosol development. 200 1.20 Hinemaiaia Tephra within White fine pumice and coarse pumice ash scattered within fine brown Pp greasy ash. 400 Pp Brown and yellow brown greasy fine ash with paleosol development. 300 Grey and greyish brown fine-med sands, planar and wavy bedded. 1000+ Grey sandy matrix diamicton, pebble-boulder sub-rounded grey clasts supported within matrix, firm, massive and unbedded. 2.90 Base into river.

Section 99 Tongariro River, T19 544423 Judges pool, exposure on true right bank of river, second terrace scarp.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

1500 Grey and black med-coarse bedded sands, fine white Taupo pumice interbedded, wavy and planar bedded on a 20-50 mm scale 400 Taupo lgnimbrite White fine-coarse pumice ash, poorlysorted and massive, sharp upper contact. 1200 Bedded coarse-med gravels, rounded grey clasts supported, with grey well sorted sand between clasts. 1000+ Grey sandy matrix diamicton, pebble-boulder sub-rounded grey clasts supported within matrix, massive and unbedded. 3.10 Base obscured.

Section Tongariro10 River,0 T19 545424 Judges Pool, True right bank of river, cliff exposure.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

2000- Taupo lgnimbrite White fine-coarse pumice ash and mixed white pumice lapilli and 30000 blocks, massive with charcoal fragments and logs near base, upper part reworked and planar and cross-bedded. 1000 Papakai Formation Yellow brown and brown fine greasy ash, with paleosol development. 400 Mangamate Tephra, Poutu Strong brown and yellow brown fine pumice lapilli with scattered grey Lapilli fine lithic lapilli, shower-bedded and ungraded. 50-100 Poronui Tephra Grey and greyish brown fine ash with pocketing fine white pumice ash. 200 Yellow brown and brown fine ash. 100-150 Grey med ash. 300 Greyish brown fineash. 1000 Grey and greyish brown fine grained diamicton, sands, silts and fine gravels cross-bedded and planar bedded. 1500 Sharp upper and lower contact, unit as above but only planar bedded. 12000 Greyish brown fine-coarse sand massive and unbedded with faint planar fabric, several units separated by 10-50 mm greyish brown silt or fine sand, fine pebbles supported within the sandy matrix. Firm and bluff forming. 16.60 Base obscured.

A 69 Section 101 Tongariro River, T19 538417 True left bank of river at cable-way across river.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

500 Brownish black fine-coarse bedded sands with scattered fine white Taupo Pumice pebbles within, recent soil development in top. 2000 Clast supported pebble, cobble and small boulder deposit, sands between clasts. Pebbles of Taupo Pumice within. 1000 Grey and greyish brown sandy matrix diamicton, with pebble-boulder sub-rounded clasts supported by matrix, clast rich unit. 3.50 Base into river.

Section 102 Tongariro River, T19 543427 True right bank of river, 100 m downstream of State Highway 1 bridge.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

350 Brownish grey and grey bedded sands cross and planar bedded on a 20 mm scale, Taupo Pumice fine pebbles interbedded. 2100 Grey rounded cobbles and pebble and sand, clasts supported, loose and unconsolidated, very rudimentary soil development in top. Highly rounded in places with zones devoid of matrix and zones with sand matrix between clasts although not supporting clasts. Taupo Pumice fine pebbles present within top of deposit. 80-85 % andesite clasts and remainder greywacke clasts. 2.45 Base into river

Section 103 Tongariro River, T19 540429 Swirl Pool, true left bank of river.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

500-1 500 Pale brown and grey fine sands and silts cross-bedded and containing Taupo Pumice fine pebbles. Alternating layers of silt and sand, bedded on a 10-50 mm scale. Present, recent soil development in top. Irregular lower contact. 0-500 Pale greyish brown and greyish brown silt and fine sand, massive and mottled with pale greyish brown colours, structureless with common root channels. 150-250 Coarse sandy matrix diamicton, pebble clasts supported within the matrix and with common scattered white fine Taupo pumice rounded pebbles, massive and unbedded. 2500+ Grey and greyish brown diamicton, in places matrix supported and in some places clast supported where its upper part is reworked. Grey sandy matrix firm where it is supporting, and grey sub-rounded and sub-angular pebble-boulder clasts, firm unit, massive with no bedding in all of the unit. 4.65 Base into river.

Section 104 Tongariro River, T19 541435 Upper Island Pool, true right bank of river.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

200 Black and brown fine sand and silt, with scattered white fine rounded Taupo Pumice lapilli, present soil development within this layer, friable and unbedded. 400 Bedded grey and greyish brown sands with scattered white fineTaupo Pumice pebbles, wavy bedded with 10-20 mm scale beds. 50 Fine gravel and sand layer with andesite, greywacke and Taupo Pumice fine-med pebbles and grey sand, clast supported. 50 Greyish brown fine sand with scattered fine Taupo Pumice fine pebbles. 50 Fine gravel and sand layer with andesite, greywacke and Taupo Pumice fine-med pebbles and grey sand, clast supported. 100 Greyish brown fine sand with scattered fine Taupo Pumice fine

A 70 pebbles, 10 mm scale wavy bedding. 1500+ Grey pebbles, cobbles and boulders and coarse sand, clast supported but with matrix of coarse sand between clasts. Sparse Taupo Pumice pebbles within unit. Andesite and greywacke clasts, unconsolidated and unbedded. 2.35 Base into river.

Section 105 Tongariro River, T19 524425 Stream running through western residential partof town , below major road culvert, highest terrace seen in town.

Unit Depth Cum. Correlation and samples Description (mm) depth (m) taken

1000 Taupo lgnimbrite White fine-coarse pumice ash with mixed white pumice lapilli and blocks, primary in places, reworked with grey sands in others. 1000 Papakai Formation and Yellow brown and brown fine ash, paleosolic and greasy, 100-150 mm Hinemaiaia Tephra zone in centre of unit with scattered, sparse, white finepumice lapilli, and coarse ash 300 2.3 Mangamate Tephra, Poutu Strong brown and yellow brown fine pumice lapilli with scattered grey Lapilli fine lithic lapilli. 20 Greyish brown fine-med ash. 100 Greyish brown coarse pumice and lithic ash and fine lapilli. 50 Yellow brown fine ash. 200 2.57 Mangamate Tephra, Te Grey, yellow brown and pale yellow fine pumice lapilli and coarse ash. Rato Lapilli 1000 Obscured. 2000 Grey and pale greyish brown fine grained diamicton. Sand and silt matrix supporting pebble clasts, planar fabric. 1000 Firm diamicton, Grey and greyish brown sandy matrix supporting clasts of pebble-cobble dominantly but also some boulders, massive and unbedded. 6.57 Base into creek.

A 71